A method and device for predicting carbon emissions during highway construction period

Abstract

The present invention relates to the technical field of carbon emission calculation, and specifically to a method and device for predicting carbon emissions during the construction period of a highway. In order to accurately predict carbon emissions, it is necessary to obtain time series data of multiple environmental factors in the local area of ​​the road section to be repaired, predict future environmental time series data, and analyze the impact of various environmental factors on construction. In order to improve the accuracy, based on the similarity of the time series data of the historical construction process of the repaired road section, the fluctuation state of the historical environmental time series data is analyzed to determine the construction influencing factors. Further analyze the similarity between the historical and predicted environmental time series data, and determine the environmental similarity value in combination with the construction influencing factors. At the same time, analyze the similarity between the historical construction plan and the construction plan and geological parameters to determine the construction similarity value. Combine the two to screen out similar reference sections, and train a neural network to improve the accuracy of the carbon emission prediction model. Finally, based on the trained neural network and the predicted environmental time series data, predict the construction plan and calculate the carbon emissions.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission calculation, and in particular to a method and device for predicting carbon emission during highway construction period. Background Art

[0002] As an important part of infrastructure construction, the carbon emissions generated during highway construction cannot be ignored. Highway construction involves complex construction processes, such as the operation of equipment, energy, and the use of building materials, which will all produce carbon emissions. In order to better monitor the carbon emissions during highway construction, carbon emission forecasting can be performed.

[0003] Traditional carbon emission prediction methods usually use simplified assumptions, such as LCA-based calculation methods. However, in the actual construction process, the construction environment, such as weather conditions, geological factors, etc., will cause changes in the specific construction process of the highway, and the corresponding carbon emissions will also change dynamically. Therefore, the simplified assumption-based LCA calculation method is difficult to apply to the dynamically changing construction environment, which will result in large errors in the final carbon emission prediction results. Summary of the invention

[0004] In order to solve the technical problem that weather conditions, geological factors, etc. will cause changes in the specific construction process of the highway, the corresponding carbon emissions will also change dynamically. Therefore, the simplified assumption-based LCA calculation method is difficult to apply to the dynamically changing construction environment, which will lead to a large error in the final carbon emission prediction result. The purpose of the present invention is to provide a method and device for predicting carbon emissions during the construction period of a highway. The technical scheme adopted is as follows:

[0005] Obtain reference environmental time series data of various environmental factors in the local area where the road section to be repaired is located, obtain various historical construction process time series data, historical environmental time series data, historical construction plans and various geological parameters of the repaired road section, and obtain the construction plan and various geological parameters of the road section to be repaired;

[0006] Predict multiple reference environmental time series data to obtain multiple predicted environmental time series data for future periods of the road section to be repaired; analyze the fluctuations of historical environmental time series data based on the similarity between historical construction process time series data, so as to determine the construction influencing factors of each environmental factor; determine the environmental similarity value based on the similarity between historical environmental time series data and predicted environmental time series data, and in combination with the construction influencing factors of environmental factors;

[0007] Analyze the similarity between the historical construction plan of the repaired road section and the construction plan of the road section to be repaired, as well as the similarity between the geological parameters of the repaired road section and the road section to be repaired, and determine the construction similarity value; comprehensively consider the environmental similarity value and the construction similarity value between the road section to be repaired and the repaired road section, and select the reference road section from all the repaired road sections; use the historical environmental time series data and the historical construction plan corresponding to the reference road section as a data set to obtain a trained neural network;

[0008] Based on the trained neural network and the predicted environmental time series data of the road section to be repaired, the predicted construction plan of the road section to be repaired is determined; according to the predicted construction plan of the road section to be repaired, the usage of raw materials and the vegetation area, the predicted carbon emissions of the road section to be repaired are determined.

[0009] Furthermore, the method for obtaining the construction influencing factors includes:

[0010] In the historical construction process time series data, the similarity of data values ​​between moments is analyzed, and then all moments are clustered to obtain clusters;

[0011] In each cluster, for any environmental factor, at all times, the difference between the maximum environmental value and the minimum environmental value corresponding to the environmental factor is taken as the fluctuation amplitude value of the environmental factor in each cluster;

[0012] The mean of the fluctuation amplitude values ​​of each environmental factor in all clusters is negatively correlated and normalized to obtain the value as the construction influencing factor of each environmental factor.

[0013] Furthermore, the method for obtaining the clustering clusters includes:

[0014] In all the historical construction process time series data of each repaired road section, all the construction process values ​​at each moment are arranged in the same arrangement mode, so as to obtain the construction process vector corresponding to each moment;

[0015] Calculate the cosine similarity between the construction process vectors at any two moments and perform negative correlation mapping to obtain the construction difference value between any two moments;

[0016] Based on the DBSCAN clustering algorithm, cluster analysis is performed on all moments to obtain cluster clusters, where the distance metric is the construction difference value between moments.

[0017] Furthermore, the method for obtaining the environment similarity value includes:

[0018] For any repaired road section, under the same environmental factors, the value after negative correlation mapping between the DTW distance between the historical environmental time series data of the repaired road section and the predicted environmental time series data of the road section to be repaired is used as the environmental change similarity value;

[0019] The product of the DTW distance between the previous repaired road section adjacent to the road section to be repaired and the road section to be repaired under each environmental factor and the construction impact factor of each environmental factor is normalized as the construction impact coefficient of each environmental factor;

[0020] For any repaired road section, the environmental change similarity values ​​of the environmental factors are weighted and averaged using the construction impact coefficient of the environmental factors, so as to obtain the environmental similarity value between the repaired road section and the road section to be repaired.

[0021] Furthermore, the method for obtaining the construction similarity value includes:

[0022] Analyze the similarity between the historical construction plan of each repaired road section and the construction plan of the road section to be repaired, and determine the first construction similarity factor;

[0023] For any repaired road section, the absolute value of the difference between the repaired road section and the road section to be repaired under the same geological parameters is calculated as the geological difference factor, and the mean of the geological difference factors under all types of geological parameters is negatively correlated and normalized to the value obtained as the second construction similarity factor between the repaired road section and the road section to be repaired;

[0024] The normalized value of the sum of the first construction similarity factor and the second construction similarity factor corresponding to each repaired road section is used as the construction similarity value between each repaired road section and the road section to be repaired.

[0025] Furthermore, the method for obtaining the first construction similarity factor includes:

[0026] The construction plan and each historical construction plan are processed using Jieba word segmentation tool to obtain vocabulary sets corresponding to each repaired road section and the road section to be repaired;

[0027] The Jaccard coefficient between the vocabulary set of the road section to be repaired and the vocabulary set of each repaired road section is calculated as the first construction similarity factor between the road section to be repaired and each repaired road section.

[0028] Furthermore, the method for obtaining the reference road section includes:

[0029] The product of the environmental similarity value and the construction similarity value corresponding to each repaired road section is normalized to obtain the value as the comprehensive similarity characteristic value between each repaired road section and the road section to be repaired;

[0030] Among all the repaired road sections, the repaired road sections whose comprehensive similarity feature values ​​are greater than a preset similarity threshold are used as reference road sections.

[0031] Furthermore, the method for obtaining the predicted carbon emissions includes:

[0032] The predicted carbon emissions are the sum of the reduction in vegetation carbon sinks, carbon emissions from material production, and carbon emissions from construction;

[0033] The formula model for the reduction of vegetation carbon sink includes: in, It indicates the reduction of vegetation carbon sink; Indicates Daily carbon sequestration per unit land area of ​​vegetation; Indicates Plant cover per unit land area, Indicates the design service life of the expressway; Indicates the number of vegetation types;

[0034] The formula model for carbon emissions from material production includes: in, Indicates the carbon emissions of material production; Indicates the predicted construction plan The first step in the production process of engineering materials Energy consumption; Indicates the predicted construction plan The first step in the production process of engineering materials Carbon emission coefficient corresponding to the energy type; Indicates the predicted construction plan The consumption of the material; Indicates the number of types of engineering materials in the predicted construction plan; Indicates the predicted construction plan The number of types of energy required in the production process of engineering materials;

[0035] The formula model of construction carbon emissions includes: in, It represents the construction carbon emission of the predicted construction scheme; Indicates the predicted construction plan Construction activity The amount of energy consumed; Indicates the predicted construction plan Construction activity Carbon emission factors of the energy sources; It represents the number of construction activities in the forecasted construction plan; Indicates the predicted construction plan The number of energy types required for each construction activity.

[0036] Furthermore, the method for obtaining the prediction environment time series data includes:

[0037] The ARIMA model is used to process the reference environmental time series data of the road section to be repaired under each environmental factor, so as to obtain the predicted environmental time series data of the preset length of each environmental factor in the future period of the road section to be repaired.

[0038] A device for predicting carbon emissions during a highway construction period includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and when the at least one instruction, at least one program, a code set or an instruction set is loaded and executed by the processor, the steps of a method for predicting carbon emissions during a highway construction period are implemented.

[0039] The present invention has the following beneficial effects:

[0040] Given that the construction environment will affect the carbon emissions during the specific construction process of the highway, the reference time series data of various environmental factors in the local area where the section to be repaired is located is obtained, and it is used as a reference to predict the environmental time series data of the future period, which can improve the accuracy of environmental data prediction. Then, the impact of each environmental factor on the construction can be further analyzed. In order to improve the accuracy of the obtained impact, the fluctuation state of the historical environmental time series data of the repaired section can be analyzed on the basis of the similarity of the historical construction process time series data of the repaired section, so as to determine the construction impact factor of each environmental factor, which is helpful to accurately identify which environmental factors have a significant impact on carbon emissions. Further, by analyzing the similarity between the historical environmental time series data and the predicted environmental time series data, and combining the construction impact factor of the environmental factor, the environmental similarity value of the repaired section and the section to be repaired can be determined, which can be used as one of the indicators for subsequent screening of reference sections for predicting carbon emissions. After determining the environmental similarity value, the similarity between the historical construction plan of the repaired section and the construction plan of the section to be repaired, as well as the similarity between the geological parameters of the repaired section and the section to be repaired, can be continued to determine the construction similarity value. Then, by combining the construction similarity value with the environmental similarity value, we screened out reference sections that are similar to the environmental status and construction status of the sections to be repaired, ensuring the accuracy and representativeness of the reference data, and trained the neural network based on various data of the reference sections, thereby improving the generalization ability and prediction accuracy of the carbon emission prediction model. Finally, based on the trained neural network and the predicted environmental time series data of the sections to be repaired, we can scientifically predict the construction plan of the sections to be repaired, and then accurately calculate the predicted carbon emissions of the sections to be repaired based on the construction plan, raw material usage and vegetation area, providing strong support for the green and low-carbon construction of highways. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A method flow chart of a method for predicting carbon emissions during highway construction period provided by one embodiment of the present invention;

[0043] Figure 2 A method flow chart of a method for obtaining construction influencing factors provided by an embodiment of the present invention;

[0044] Figure 3 A method flow chart of a method for obtaining construction similarity values ​​provided by one embodiment of the present invention;

[0045] Figure 4 A schematic diagram of the structure of a device for predicting carbon emissions during highway construction period provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a method and device for predicting carbon emissions during the construction period of a highway proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0047] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0048] The specific scheme of a method and device for predicting carbon emissions during highway construction period provided by the present invention is described in detail below with reference to the accompanying drawings.

[0049] See also Figure 1 , which shows a method flow chart of a method for carbon emission during highway construction period provided by an embodiment of the present invention, the method comprising the following steps:

[0050] Step S1: Obtain reference environmental time series data of various environmental factors in the local area where the road section to be repaired is located, obtain various historical construction process time series data, historical environmental time series data, historical construction plans and various geological parameters of the repaired road section, and obtain the construction plan and various geological parameters of the road section to be repaired.

[0051] The carbon emissions of a highway construction project refer to the total amount of carbon dioxide and other greenhouse gases generated during the construction process, which includes the sum of the vegetation carbon sink reduced by land acquisition, the carbon emissions from material production and the carbon emissions from the construction process. The carbon emissions from material production are mainly due to the fact that highway construction projects require a large amount of materials and energy, such as cement, asphalt, steel and electricity. The production, transportation and use of these resources will be accompanied by certain carbon emissions. The carbon emissions from the construction process are mainly due to the fact that the mechanical equipment, vehicles and power generation equipment used all require energy supply, and the burning of fuel or the use of electricity will produce corresponding carbon emissions. Therefore, when predicting carbon emissions during the highway construction period, it is usually necessary to calculate the above three types of carbon emissions of the section to be constructed. For the carbon emissions from material production and the carbon emissions from the construction process, they are mainly related to the specific construction plan. Therefore, the main purpose of the present invention is to obtain a more accurate predicted construction plan for the section to be constructed. The acquisition of the predicted construction plan can rely on the data of a reference section that is relatively similar to the various states of the section to be constructed for prediction.

[0052] At the same time, because the construction environment, such as environmental factors, temperature, humidity, wind speed, etc., and geological parameters, such as rock layer inclination, compressive strength, etc., will have an impact on the actual construction plan, in the embodiment of the present invention, the reference environment time series data of various environmental factors in the local area where the road section to be repaired is located is obtained, and the reference environment time series data can be used to predict the predicted environment time series data of the road section to be repaired in the subsequent process; various historical construction process time series data, historical environmental time series data, historical construction plans and various geological parameters of the repaired road section are also obtained, as well as the construction plan of the road section to be repaired and various geological parameters. Among them, the historical environmental time series data and the historical construction process time series data can be used to analyze the impact of environmental factors on the construction process, and the historical construction plan and construction plan, as well as the geological parameters can be used to compare the construction status between the road section to be repaired and the repaired road section.

[0053] In this embodiment of the present invention, environmental factors can be set to humidity, temperature, wind speed, wind force, etc. The specific method for obtaining historical environmental time series data can be to install corresponding sensors at appropriate locations on the construction site for acquisition, and the main purpose of the reference environmental time series data is to predict the predicted environmental time series data of the road section to be repaired in the future period, so the reference time series data of various environmental factors in the county / city where the road section to be repaired is located can be obtained through the meteorological department; the types of historical construction process time series data may include power, vibration, pressure, etc. of various construction facilities, and the specific acquisition method can also be to install sensors at appropriate locations of the construction facilities for acquisition; the types of geological parameters include rock layer inclination, compressive strength, etc., and the specific acquisition method can be obtained by geological radar; the historical construction plan of the repaired road section and the construction plan of the road section to be repaired can be obtained through historical construction records and construction plans, among which the contents of the historical construction records and construction plans should include: construction methods, material usage in construction activities, usage of construction facilities, length of the construction section, construction days and other related data.

[0054] It should be noted that the length of the reference environment time series data can be set to one week from the current moment; the historical environment time series data and the historical construction process time series data should be data from the same time period, and the length is determined by the specific construction duration of the repaired road section; the sampling time interval of the time series data is set to 1 second, and the specific interval can be adjusted according to the implementation scenario and is not limited here.

[0055] Step S2: Predict a variety of reference environmental time series data to obtain a variety of predicted environmental time series data for future periods of the road section to be repaired; based on the similarity between the historical construction process time series data, analyze the fluctuations of the historical environmental time series data to determine the construction influencing factors of each environmental factor; based on the similarity between the historical environmental time series data and the predicted environmental time series data, and combined with the construction influencing factors of the environmental factors, determine the environmental similarity value.

[0056] Highway construction is a long and complex process, and the specific construction conditions will also be affected by environmental factors, so its carbon emissions are affected by a variety of environmental factors. For example, as the weather changes, when building a highway, it is necessary to adjust the construction machinery used according to the actual environment and make corresponding response plans; however, these targeted adjustments use unnecessary machinery and build unnecessary response measures. Therefore, compared with the carbon emissions originally planned, some additional carbon emissions are added; and if the same weather is encountered, the same or similar avoidance measures will be taken. Therefore, by predicting the environmental conditions of the section to be repaired in the future period, a variety of predicted environmental time series data for the section to be repaired in the future period are obtained, which can more accurately evaluate the construction situation in the subsequent process, thereby more accurately predicting carbon emissions.

[0057] Then, by analyzing the similarity between the historical construction process time series data and analyzing the fluctuation of environmental factors under the same construction conditions, the influence of each environmental factor on the construction situation can be calculated. Then, based on the similarity between the historical environmental time series data of the repaired road section and the predicted environmental time series data of the road section to be repaired, and combined with the construction impact factor of the environmental factors on the construction situation, the environmental similarity value between the road section to be repaired and the repaired road section can be obtained, which can reflect the similarity of the environmental status between the road section to be repaired and the repaired road section, and can therefore be used as an indicator for subsequent screening of reference sections, which helps to more accurately predict the construction plan of the road section to be repaired.

[0058] There are traces of weather changes. In an area, the weather in the previous period usually affects the weather conditions in the next period. Therefore, the environmental conditions of the section to be repaired in the future period can be predicted based on the environmental conditions of the local area where the section to be repaired is located. That is, based on multiple reference environmental time series data, multiple predicted environmental time series data for the future period of the section to be repaired can be obtained.

[0059] Preferably, in one embodiment of the present invention, the method for obtaining the prediction environment time series data includes:

[0060] The ARIMA model is used to process the reference environmental time series data of the road section to be repaired under each environmental factor, so as to obtain the predicted environmental time series data of the preset length of each environmental factor in the future period of the road section to be repaired.

[0061] It should be noted that the ARIMA model is a well-known technology and the specific process will not be described in detail here; the preset length is set to 5 days, and the specific length can be adjusted according to actual conditions and is not limited here.

[0062] Changes in weather, that is, changes in environmental factors, will lead to changes in construction conditions, so it is necessary to analyze the construction impact factors of environmental factors on construction conditions. In order to obtain more accurate construction impact factors, the similarity values ​​between the historical construction process time series data can be analyzed. On the basis of the consistency of the construction process, the fluctuation characteristics of the historical environmental time series data can be analyzed to capture the changes in environmental factors, so as to more accurately quantify the construction impact factors of environmental factors on construction.

[0063] Preferably, in one embodiment of the present invention, the method for obtaining the construction influencing factor includes:

[0064] See also Figure 2 , which shows a method flow chart of a method for obtaining construction influencing factors in one embodiment of the present invention, the method comprising the following steps:

[0065] Step S201: In the historical construction process time series data, the similarity of data values ​​between moments is analyzed, so as to perform cluster analysis on all moments and obtain clusters.

[0066] Since the historical construction process time series data and historical environmental time series data of the same repaired road section are in the same period, in all types of historical construction process time series data of each repaired road section, if the type of construction process data is 3, then each moment corresponds to three construction process values; all construction process values ​​at each moment are arranged in the same arrangement method to obtain the construction process vector corresponding to each moment. For example, the construction process data are recorded as construction process data 1, construction process data 2 and construction process data 3 respectively, then the three construction process values ​​at each moment can be arranged in the manner of (construction process data 1, construction process data 2, construction process data 3) to obtain the construction process vector corresponding to each moment.

[0067] Then calculate the cosine similarity between the construction process vectors at any two moments. The larger the cosine similarity, the more similar the construction process vectors at the two moments are, that is, the construction process data is relatively consistent. Conversely, the smaller the cosine similarity, the greater the difference between the construction process vectors at the two moments, that is, the consistency of the construction process data is low. Therefore, the obtained cosine similarity can be negatively correlated to achieve logical relationship correction, thereby obtaining the construction difference value between any two moments. At this time, the smaller the construction difference value, the more consistent the construction process data at the two moments are, that is, the more they should belong to the same type of construction situation. Since the value range of cosine similarity is [-1,1], the negative correlation mapping here can be used as formula , where x represents the independent variable.

[0068] Based on the above process, the construction difference value between any two moments can be obtained in all moments. Finally, all moments can be clustered based on the DBSCAN clustering algorithm to obtain clusters, where the distance metric is the construction difference value between the moments. After clustering, the moments in each cluster can have a relatively consistent construction situation.

[0069] It should be noted that, in an embodiment of the present invention, when clustering is performed using the DBSCAN clustering algorithm, the clustering radius is set to 0.3, and the minimum number of points within the radius is set to 12, and both can be adjusted according to the implementation scenario, and the specific values ​​are not limited; the DBSCAN clustering algorithm is a well-known technology, and the specific process is not described here; in other embodiments of the present invention, the clustering method may also adopt a K-means clustering algorithm, and the optimal K value is obtained based on the elbow method, and then the optimal K value is used to perform K-means clustering on all moments, and the clustering results can also be obtained, and the moments in each clustering cluster also have a relatively consistent construction situation, wherein the distance metric is the construction difference value between the moments, and the K-means clustering algorithm and the elbow method are both well-known technologies, and the specific process is not described here.

[0070] Step S202: Analyze the fluctuation of each environmental factor in all clusters, so as to obtain the construction influencing factor of each environmental factor.

[0071] The construction process data at all times in each cluster have relatively consistent characteristics. Therefore, if a certain environmental factor has a greater impact on the construction process, then the construction process is more sensitive to changes in the value of this environmental factor, and the environmental values ​​of this environmental factor at all times in each cluster should be relatively consistent; conversely, if a certain environmental factor has a smaller impact on the construction process, then it means that the construction process is less sensitive to changes in the value of this environmental factor, so the distribution of environmental values ​​in the cluster may show greater randomness or volatility.

[0072] Therefore, in each cluster, for any environmental factor, at all times, the difference between the maximum environmental value and the minimum environmental value corresponding to this environmental factor is taken as the fluctuation amplitude value of this environmental factor in each cluster. The larger the fluctuation amplitude value, the greater the randomness or volatility of the environmental value of this environmental factor in each cluster. Therefore, it can be considered that the impact of this environmental factor on the construction process data is relatively low.

[0073] Finally, the mean of the fluctuation amplitude of each environmental factor in all clusters is calculated. Similarly, the larger the mean, the lower the average impact of this environmental factor on the construction process data in all clusters. In order to achieve logical relationship correction, the mean is negatively correlated and normalized to obtain the construction impact factor of each environmental factor. At this time, the larger the construction impact factor of a certain environmental factor, the greater the impact of this environmental factor on the construction process. The negative correlation mapping and normalization method can be used ,in, It represents an exponential function with the natural constant e as the base, and x represents the independent variable.

[0074] In other embodiments of the present invention, the construction influencing factor of each environmental factor may also be determined by the following method:

[0075] Given that the variance of a set of data can reflect the degree of discreteness of the data, that is, its volatility, in each cluster, for any environmental factor, the variance of the environmental values ​​at all times under this environmental factor is calculated. The larger the variance, the greater the fluctuation of the environmental value of this environmental factor, that is, the environmental value of this environmental factor in each cluster shows greater randomness or volatility. Therefore, it can be considered that this environmental factor has a lower degree of influence on the construction process data.

[0076] Finally, the mean of the variance of each environmental factor in all clusters is calculated. Similarly, the larger the mean, the lower the average impact of this environmental factor on the construction process data in all clusters. In order to achieve logical relationship correction, the mean is negatively correlated and normalized to obtain the construction impact factor of each environmental factor. At this time, the larger the construction impact factor of a certain environmental factor, the greater the impact of this environmental factor on the construction process.

[0077] In the above steps, the construction impact factor corresponding to each environmental factor can be calculated, and through prediction, the predicted environmental time series data of the future period of the section to be repaired can be obtained. Therefore, the construction impact factor of the environmental factor can be combined, and the similarity between the historical environmental time series data of the repaired section and the predicted environmental time series data of the section to be repaired can be analyzed, so as to determine the environmental similarity value between the repaired section and the section to be repaired, so as to prepare for the subsequent screening of suitable reference sections.

[0078] Preferably, in one embodiment of the present invention, the method for obtaining the environment similarity value includes:

[0079] For any repaired road section, under the same environmental factors, the DTW distance between the historical environmental time series data of the repaired road section and the predicted environmental time series data of the road section to be repaired is calculated. The smaller the DTW distance, the more similar the historical environmental time series data and the predicted environmental time series data are. Therefore, the DTW distance is negatively correlated and mapped to achieve logical relationship correction, thereby obtaining the environmental change similarity value. At this time, the larger the environmental change similarity value, the higher the degree of consistency between the historical environmental time series data of the repaired road section and the predicted environmental time series data of the road section to be repaired. The negative correlation mapping method can be used ,in, It represents an exponential function with the natural constant e as the base, and x represents the independent variable.

[0080] Since the construction of expressways is a relatively long process, the entire process is usually divided into many stages to construct the corresponding construction process. However, in a certain stage, a device for countermeasures is built under the influence of weather, which means that in the next stage, when the weather or environment is the same, the measure does not need to be repeated. Therefore, when the weather or environmental factors are the same, the weight of the corresponding environmental factors needs to be reduced. For the section to be repaired, the environmental factors of the previous repaired section adjacent to it will have a greater impact on its construction status and will be more referenced. In addition, the higher the similarity of a certain environmental factor, the lower the weight of the environmental factor should be. The smaller the DTW distance, the higher the similarity. Therefore, the product of the DTW distance between the previous repaired section adjacent to the section to be repaired and the section to be repaired under each environmental factor and the construction influence factor of each environmental factor is normalized as the construction influence coefficient of each environmental factor. At this time, the construction influence coefficient of each environmental factor takes into account the influence of the construction situation caused by the environmental factors of the previous repaired section adjacent to the section to be repaired on the construction status of the section to be repaired, and thus will be more accurate. Normalization is a technical means well known to those skilled in the art, and the normalization function may be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0081] Finally, for any repaired road section, the construction impact coefficient of the environmental factor is used to perform weighted averaging on the environmental change similarity value of the environmental factor. Specifically, the construction impact coefficient of each environmental factor is multiplied by the environmental change similarity value corresponding to the repaired road section under each environmental factor to obtain the similarity coefficient under each environmental factor. The larger the similarity coefficient, the greater the impact of the environmental factor on the construction, and the higher the similarity of the environmental values ​​between the repaired road section and the road section to be repaired under the environmental factor. The average of the similarity coefficients under all environmental factors is used as the environmental similarity value between the repaired road section and the road section to be repaired. The larger the environmental similarity value, the higher the similarity between the environmental state of the repaired road section and the predicted environmental state of the road section to be repaired, and the higher the possibility that the repaired road section will be selected as a reference road section in the subsequent process.

[0082] It should be noted that the calculation of the DTW distance is a well-known technology, and the specific process will not be described in detail here.

[0083] Step S3: Analyze the similarity between the historical construction plan of the repaired road section and the construction plan of the road section to be repaired, as well as the similarity between the geological parameters of the repaired road section and the road section to be repaired, and determine the construction similarity value; comprehensively consider the environmental similarity value and the construction similarity value between the road section to be repaired and the repaired road section, and select the reference road section from all the repaired road sections; use the historical environmental time series data and historical construction plan corresponding to the reference road section as data sets to obtain a trained neural network.

[0084] By analyzing the similarities between the historical construction plans of the repaired sections and the construction plans of the sections to be repaired, as well as the similarities between their geological parameters, the construction similarity values ​​between the repaired sections and the sections to be repaired can be quantified, which helps to identify the repaired sections that are closest to the sections to be repaired in terms of construction conditions and requirements. Then, by combining the environmental similarity values ​​and construction similarity values, the reference sections that are most similar to the sections to be repaired can be screened out from all the repaired sections. These reference sections will serve as the basis for subsequent analysis and prediction. The historical environmental time series data and historical construction plans corresponding to the selected reference sections are used as data sets to provide rich historical information similar to the sections to be repaired for training neural networks. This helps the neural network learn the construction and environmental characteristics related to the sections to be repaired, thereby improving the accuracy of subsequent predictions.

[0085] If the geological conditions and location conditions of the highway are different, then there will be differences in the construction plans. For example, there will be great differences between the construction plans for bridges and tunnels, and there will be differences in the construction facilities and methods used. Therefore, by analyzing the similarities between the historical construction plans of the sections that have been built and the construction plans of the sections to be built, as well as the similarities between the geological parameters of the sections that have been built and the sections that are to be built, the construction similarity value between the sections that have been built and the sections that are to be built can be quantified to describe the construction similarities of the two sections.

[0086] Preferably, in one embodiment of the present invention, the method for obtaining the construction similarity value includes:

[0087] See also Figure 3 , which shows a method flow chart of a method for obtaining construction similarity values ​​in one embodiment of the present invention, the method comprising the following steps:

[0088] Step S301: Analyze the similarity between the historical construction plan of each repaired road section and the construction plan of the road section to be repaired, and determine the first construction similarity factor.

[0089] The construction plan and each historical construction plan are processed using the Jieba word segmentation tool to obtain the vocabulary sets corresponding to each repaired road section and the road section to be repaired.

[0090] Then the Jaccard coefficient between the vocabulary set of the road section to be repaired and the vocabulary set of each repaired road section is calculated as the first construction similarity factor between the road section to be repaired and each repaired road section; when the first construction similarity factor between a certain repaired road section and the road section to be repaired is larger, it indicates that the construction plan of the road section to be repaired has a higher consistency with the historical construction plan of the repaired road section, then the reference value of the repaired road section is higher.

[0091] It should be noted that the Jieba word segmentation tool and the calculation of the Jaccard coefficient are both well-known technologies, and the specific process will not be described here.

[0092] Step S302: Analyze the similarity between the geological parameters of each repaired road section and the geological parameters of the road section to be repaired, and determine a second construction similarity factor.

[0093] For any repaired road section, calculate the absolute value of the difference between the repaired road section and the road section to be repaired under the same geological parameters as the geological difference factor. The larger the geological difference factor, the greater the difference in geological conditions between the repaired road section and the road section to be repaired. In this case, the construction conditions and construction plans will be very different to a large extent, and the reference value needs to be reduced.

[0094] Calculate the mean of the geological difference factors of the repaired road section and the road section to be repaired under all types of geological parameters. Similarly, the larger the mean, the greater the difference in geological conditions, and the reference value of the repaired road section needs to be reduced. Therefore, the mean is negatively correlated and normalized to achieve logical relationship correction, thereby obtaining the second construction similarity factor of the repaired road section and the road section to be repaired. At this time, the larger the second construction similarity factor, the more similar the geological conditions of the repaired road section and the road section to be repaired are, and the reference value is higher. The negative correlation mapping and normalization method can be used ,in, It represents an exponential function with the natural constant e as the base, and x represents the independent variable.

[0095] Step S303: The first construction similarity factor and the second construction similarity factor between each repaired road section and the road section to be repaired are integrated to obtain a construction similarity value between each repaired road section and the road section to be repaired.

[0096] Based on the analysis of the above steps, it can be known that the larger the first construction similarity factor between a certain repaired road section and a road section to be repaired, the higher the reference value of the repaired road section; the larger the second construction similarity factor, the higher the reference value of the repaired road section; so here, the sum of the first construction similarity factor and the second construction similarity factor corresponding to each repaired road section is normalized as the construction similarity value between each repaired road section and the road section to be repaired. At this time, the larger the construction similarity value, the greater the reference value of the construction plan of the repaired road section to the road section to be repaired. Normalization is a technical means well known to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0097] After obtaining the construction similarity value between each repaired road section and the road section to be repaired, the construction similarity value and the environmental similarity value can be integrated to obtain a comprehensive index, that is, the comprehensive similarity characteristic value between the road section to be repaired and each repaired road section, so as to more comprehensively evaluate the similarity of the environment and construction conditions between each repaired road section and the road section to be repaired, so as to facilitate the screening of reference sections with more reference value and representativeness.

[0098] Preferably, in one embodiment of the present invention, the reference section screening process includes:

[0099] Based on the analysis of the above steps, it can be seen that the greater the environmental similarity value between a repaired road section and a road section to be repaired, the higher the degree of similarity between the environmental state of the repaired road section and the predicted environmental state of the road section to be repaired, and the reference value of the repaired road section is higher; similarly, if the construction similarity value between the repaired road section and the road section to be repaired is greater, it also indicates that the construction plan of the repaired road section has a greater reference value to the road section to be repaired.

[0100] Therefore, the product of the environmental similarity value and the construction similarity value corresponding to each repaired road section is normalized as the comprehensive similarity characteristic value between each repaired road section and the road section to be repaired. At this time, the larger the comprehensive similarity characteristic value of a certain repaired road section is, the higher the reference value of the repaired road section for the construction process of the road section to be repaired. Normalization is a technical means well known to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0101] Finally, among all the repaired road sections, the repaired road sections whose comprehensive similarity feature values ​​are greater than the preset similarity threshold are used as reference road sections.

[0102] It should be noted that, in this embodiment of the present invention, the preset similarity threshold is set to 0.65, and the specific value can be adjusted according to the actual scenario and is not limited here.

[0103] At this point, all reference sections can be screened out, and then the neural network can be trained based on the historical environmental time series data and historical construction plans corresponding to the reference sections, thereby obtaining a trained neural network. Since the training process of the neural network is a well-known technology, in the embodiment of the present invention, only a brief description of the training process of the neural network is given: among all the reference sections, they are divided into a training set and a validation set in a ratio of 7:3. The training set is used in the training process of the neural network to learn the mapping relationship between input data and output; the validation set is used to evaluate the performance of the model during the training process, so as to adjust parameters and select models to prevent overfitting. The historical environmental time series data of the reference sections in the training set is used as the input of the neural network RNN ​​model, and the output is the historical construction plan. The gradient descent method is used for training until the loss function converges to complete the training process of the neural network, wherein the loss function adopts the mean square error function.

[0104] Step S4: Determine the predicted construction plan of the road section to be repaired based on the trained neural network and the predicted environmental time series data of the road section to be repaired; determine the predicted carbon emissions of the road section to be repaired according to the predicted construction plan of the road section to be repaired, the usage of raw materials and the vegetation area.

[0105] In the above steps, the repaired road sections that are relatively consistent with the construction and environmental factors of the road section to be repaired are selected as reference sections, and then the neural network is trained based on the data of the reference sections. Therefore, the predicted environmental time series data of the road section to be repaired is input into the trained neural network model, and a predicted construction plan for the road section to be repaired can be automatically and quickly generated. The predicted construction plan can comprehensively consider environmental factors and the experience of historical construction plans to provide more reasonable construction guidance for the construction team. Therefore, on the basis of the predicted construction plan, combined with the carbon emissions of material production (such as steel consumption, cement consumption, asphalt consumption, iron consumption, etc.) and the reduction of vegetation carbon sinks, the predicted carbon emissions of the road section can be further obtained. The prediction of carbon emissions helps to evaluate the impact of construction activities on the environment and provide data support for formulating environmental protection measures and reducing carbon emissions.

[0106] Preferably, a method for obtaining predicted carbon emissions in one embodiment of the present invention includes:

[0107] The predicted carbon emissions are the sum of the reduction in vegetation carbon sinks, carbon emissions from material production, and carbon emissions from construction;

[0108] In the early stage of highway construction, the site needs to be leveled, and one of the tasks is land acquisition. Ground vegetation absorbs carbon dioxide during its life cycle. Therefore, land acquisition causes this part of carbon dioxide that should have been absorbed to be exposed to the air, increasing carbon emissions during the highway construction period. The formula model for the reduction of vegetation carbon sink includes: in, It indicates the reduction of vegetation carbon sink; Indicates Daily carbon sequestration per unit land area of ​​vegetation; Indicates Plant cover per unit land area, Indicates the design service life of the expressway; Indicates the number of vegetation types;

[0109] The carbon emissions from the material production process of highways are mainly concentrated in the production and use of building materials such as steel bars, cement, asphalt, etc. The formula model for carbon emissions from material production includes: in, Indicates the carbon emissions of material production; Indicates the predicted construction plan The first step in the production process of engineering materials Energy consumption; Indicates the predicted construction plan The first step in the production process of engineering materials Carbon emission coefficient corresponding to the energy type; Indicates the predicted construction plan The consumption of the material; Indicates the number of types of engineering materials in the predicted construction plan; Indicates the predicted construction plan The number of types of energy required in the production process of engineering materials;

[0110] During the construction of expressways, some carbon emissions also come from carbon emissions generated by energy consumption in specific construction activities. The formula model for construction carbon emissions includes: in, It represents the construction carbon emission of the predicted construction scheme; Indicates the predicted construction plan Construction activity The amount of energy consumed; Indicates the predicted construction plan Construction activity Carbon emission factors of the energy sources; It represents the number of construction activities in the forecasted construction plan; Indicates the predicted construction plan The number of energy types required for each construction activity.

[0111] At this point, through the above process, the predicted carbon emissions of the road sections to be repaired during the highway construction period can be obtained, so that the predicted carbon emissions can be monitored.

[0112] In summary, given that the construction environment will affect the carbon emissions during the specific construction process of the highway, obtaining the reference time series data of various environmental factors in the local area where the section to be repaired is located and using it as a reference to predict the environmental time series data for future periods can improve the accuracy of environmental data prediction. Then, the impact of each environmental factor on the construction can be analyzed. In order to improve the accuracy of the obtained impact, the fluctuation state of the historical environmental time series data of the repaired section can be analyzed on the basis of the similarity of the historical construction process time series data of the repaired section, so as to determine the construction impact factor of each environmental factor, which is helpful to accurately identify which environmental factors have a significant impact on carbon emissions. Further, by analyzing the similarity between the historical environmental time series data and the predicted environmental time series data, and combining the construction impact factor of the environmental factor, the environmental similarity value of the repaired section and the section to be repaired can be determined, which can be used as one of the indicators for subsequent screening of reference sections for predicting carbon emissions. After determining the environmental similarity value, the similarity between the historical construction plan of the repaired section and the construction plan of the section to be repaired, as well as the similarity between the geological parameters of the repaired section and the section to be repaired, can be analyzed to determine the construction similarity value. Then, by combining the construction similarity value with the environmental similarity value, we screened out reference sections that are similar to the environmental status and construction status of the sections to be repaired, ensuring the accuracy and representativeness of the reference data, and trained the neural network based on various data of the reference sections, thereby improving the generalization ability and prediction accuracy of the carbon emission prediction model. Finally, based on the trained neural network and the predicted environmental time series data of the sections to be repaired, we can scientifically predict the construction plan of the sections to be repaired, and then accurately calculate the predicted carbon emissions of the sections to be repaired based on the construction plan, raw material usage and vegetation area, providing strong support for the green and low-carbon construction of highways.

[0113] The embodiment of the present invention also provides a device for predicting carbon emissions during highway construction period. Figure 4 , which shows a schematic diagram of the device structure, including a processor 400, a memory 401, a bus 402 and a communication interface 403, wherein the processor 400, the communication interface 403 and the memory 401 are connected via the bus 402; wherein the memory 401 may include a high-speed random access memory, the bus 402 may be an ISA bus, a PCI bus or an EISA bus, etc., and the processor 400 may be an integrated circuit chip with signal processing capabilities; the memory 401 stores at least one instruction, at least one program, a code set or an instruction set, and when the at least one instruction, at least one program, a code set or an instruction set is loaded and executed by the processor, a step in a method for predicting carbon emissions during the construction period of a highway is implemented.

[0114] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0115] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for predicting carbon emissions during highway construction, characterized in that: The method comprises: Obtain reference environmental time series data of various environmental factors in the local area where the road section to be repaired is located, obtain various historical construction process time series data, historical environmental time series data, historical construction plans and various geological parameters of the repaired road section, and obtain the construction plan and various geological parameters of the road section to be repaired; Predict multiple reference environmental time series data to obtain multiple predicted environmental time series data for future periods of the road section to be repaired; analyze the fluctuations of historical environmental time series data based on the similarity between historical construction process time series data, so as to determine the construction influencing factors of each environmental factor; determine the environmental similarity value based on the similarity between historical environmental time series data and predicted environmental time series data, and in combination with the construction influencing factors of environmental factors; The method for obtaining the environment similarity value includes: For any repaired road section, under the same environmental factors, the value after negative correlation mapping between the DTW distance between the historical environmental time series data of the repaired road section and the predicted environmental time series data of the road section to be repaired is used as the environmental change similarity value; The product of the DTW distance between the previous repaired road section adjacent to the road section to be repaired and the road section to be repaired under each environmental factor and the construction impact factor of each environmental factor is normalized as the construction impact coefficient of each environmental factor; For any repaired road section, the environmental change similarity values ​​of the environmental factors are weighted and averaged using the construction impact coefficient of the environmental factors, thereby obtaining the environmental similarity value between the repaired road section and the road section to be repaired; Analyze the similarity between the historical construction plan of the repaired road section and the construction plan of the road section to be repaired, as well as the similarity between the geological parameters of the repaired road section and the road section to be repaired, and determine the construction similarity value; comprehensively consider the environmental similarity value and the construction similarity value between the road section to be repaired and the repaired road section, and select the reference road section from all the repaired road sections; use the historical environmental time series data and the historical construction plan corresponding to the reference road section as a data set to obtain a trained neural network; Based on the trained neural network and the predicted environmental time series data of the road section to be repaired, the predicted construction plan of the road section to be repaired is determined; according to the predicted construction plan of the road section to be repaired, the usage of raw materials and the vegetation area, the predicted carbon emissions of the road section to be repaired are determined.

2. A method for predicting carbon emissions during highway construction period according to claim 1, characterized in that: The method for obtaining the construction influencing factor includes: In the historical construction process time series data, the similarity of data values ​​between moments is analyzed, and then all moments are clustered to obtain clusters; In each cluster, for any environmental factor, at all times, the difference between the maximum environmental value and the minimum environmental value corresponding to the environmental factor is taken as the fluctuation amplitude value of the environmental factor in each cluster; The mean of the fluctuation amplitude values ​​of each environmental factor in all clusters is negatively correlated and normalized to obtain the value as the construction influencing factor of each environmental factor.

3. A method for predicting carbon emissions during highway construction period according to claim 2, characterized in that: The method for obtaining the clusters includes: In all the historical construction process time series data of each repaired road section, all the construction process values ​​at each moment are arranged in the same arrangement mode, so as to obtain the construction process vector corresponding to each moment; Calculate the cosine similarity between the construction process vectors at any two moments and perform negative correlation mapping to obtain the construction difference value between any two moments; Based on the DBSCAN clustering algorithm, cluster analysis is performed on all moments to obtain cluster clusters, where the distance metric is the construction difference value between moments.

4. The method for predicting carbon emissions during highway construction period according to claim 1 is characterized in that: The method for obtaining the construction similarity value includes: Analyze the similarity between the historical construction plan of each repaired road section and the construction plan of the road section to be repaired, and determine the first construction similarity factor; For any repaired road section, the absolute value of the difference between the repaired road section and the road section to be repaired under the same geological parameters is calculated as the geological difference factor, and the mean of the geological difference factors under all types of geological parameters is negatively correlated and normalized to the value obtained as the second construction similarity factor between the repaired road section and the road section to be repaired; The normalized value of the sum of the first construction similarity factor and the second construction similarity factor corresponding to each repaired road section is used as the construction similarity value between each repaired road section and the road section to be repaired.

5. A method for predicting carbon emissions during highway construction period according to claim 4, characterized in that: The method for obtaining the first construction similarity factor includes: The construction plan and each historical construction plan are processed using Jieba word segmentation tool to obtain vocabulary sets corresponding to each repaired road section and the road section to be repaired; The Jaccard coefficient between the vocabulary set of the road section to be repaired and the vocabulary set of each repaired road section is calculated as the first construction similarity factor between the road section to be repaired and each repaired road section.

6. A method for predicting carbon emissions during highway construction period according to claim 1, characterized in that: The method for obtaining the reference road section includes: The product of the environmental similarity value and the construction similarity value corresponding to each repaired road section is normalized and used as the comprehensive similarity characteristic value between each repaired road section and the road section to be repaired; Among all the repaired road sections, the repaired road sections whose comprehensive similarity feature values ​​are greater than a preset similarity threshold are used as reference road sections.

7. A method for predicting carbon emissions during highway construction period according to claim 1, characterized in that: The method for obtaining the predicted carbon emissions includes: The predicted carbon emissions are the sum of the reduction in vegetation carbon sinks, carbon emissions from material production, and carbon emissions from construction; The formula model for the reduction of vegetation carbon sink includes: in, It represents the reduction of vegetation carbon sink; Indicates Daily carbon sequestration per unit land area of ​​vegetation; Indicates Plant cover per unit land area, Indicates the design service life of the expressway; Indicates the number of vegetation types; The formula model for carbon emissions from material production includes: in, Indicates the carbon emissions of material production; Indicates the predicted construction plan The first step in the production process of engineering materials Energy consumption; Indicates the predicted construction plan The first step in the production process of engineering materials Carbon emission coefficient corresponding to the energy type; Indicates the predicted construction plan The consumption of the material; Indicates the number of types of engineering materials in the predicted construction plan; Indicates the predicted construction plan The number of types of energy required in the production process of engineering materials; The formula model of construction carbon emissions includes: in, It represents the construction carbon emission of the predicted construction scheme; Indicates the predicted construction plan Construction activity The amount of energy consumed; Indicates the predicted construction plan Construction activity Carbon emission factors of the energy sources; It represents the number of construction activities in the forecasted construction plan; Indicates the predicted construction plan The number of energy types required for each construction activity.

8. The method for predicting carbon emissions during highway construction period according to claim 1 is characterized in that: The method for obtaining the prediction environment time series data includes: The ARIMA model is used to process the reference environmental time series data of the road section to be repaired under each environmental factor, so as to obtain the predicted environmental time series data of the preset length of each environmental factor in the future period of the road section to be repaired.

9. A device for predicting carbon emissions during highway construction, characterized in that: It includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and when the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor, the steps of a method for predicting carbon emissions during the construction period of a highway as described in any one of claims 1-8 are implemented.

Images