A carbon emission monitoring system and method for green transportation in railway engineering

Abstract

The present application relates to the field of carbon emission monitoring technology, and specifically discloses a carbon emission monitoring system and method for green transportation in railway engineering. The system first obtains the actual power consumption of the train during operation from the train traction power supply system based on a predetermined time period, and simultaneously obtains the grid carbon emission factor corresponding to the train operation area. Based on this, the carbon emissions of the train in each time period are calculated. Then, high-carbon emission data items are further screened out, and the train operation status information in each high-carbon emission time period is retrieved. Then, based on deep learning technology, deep feature extraction and cluster analysis are performed on the train operation status information in each high-carbon emission time period to identify the main influencing factors leading to high carbon emissions of the train. The present application can realize refined attribution analysis of train carbon emissions, timely discover and warn potential high-carbon emission events, and thus provide a scientific basis for formulating targeted energy conservation and emission reduction strategies.

Description

Technical Field

[0001] The present application relates to the technical field of carbon emission monitoring, and more specifically, to a carbon emission monitoring system and method for green transportation in railway engineering. Background Art

[0002] As a relatively environmentally friendly mode of transportation, railway transportation has significant advantages in energy conservation and emission reduction. Currently, with the expansion of the railway network and the increase in train operation frequency, it has become particularly important to understand and optimize carbon emissions during train operation.

[0003] Currently, railway system carbon emissions monitoring primarily relies on static models based on energy consumption estimates, often employing fixed carbon emission coefficients (such as the national average emission factor) to estimate carbon emissions. This approach ignores the spatiotemporal variations in regional grid energy structures, fails to trace the coupling mechanisms of multi-dimensional factors behind high-carbon emission events, and lacks the ability to dynamically attribute high-carbon emission scenarios. For example, train driving behavior (such as sudden acceleration), line conditions (such as sections with sudden grade changes), and grid power supply characteristics (such as the grid energy mix) all have a synergistic impact on carbon emissions. Traditional carbon emission monitoring methods often struggle to dynamically monitor and analyze train operation carbon emissions, resulting in a lack of clear direction when formulating energy-saving and emission-reduction measures.

[0004] Therefore, it is necessary to provide an optimized carbon emission monitoring system and method for green transportation in railway engineering to solve the above technical problems. Summary of the Invention

[0005] In order to solve the above technical problems, this application is proposed.

[0006] According to one aspect of the present application, a carbon emission monitoring method for green transportation in railway engineering is provided, which comprises:

[0007] Acquiring actual power consumption during train operation from the train traction power supply system based on a predetermined time period to obtain a time series of actual power consumption, and acquiring a power grid carbon emission factor corresponding to the train operation area from the power grid energy management system based on the predetermined time period to obtain a time series of power grid carbon emission factors;

[0008] Calculating a time series of train carbon emissions based on the time series of the actual power consumption and the time series of the grid carbon emission factor;

[0009] Marking data items in the time series of carbon emissions of the train that are greater than a preset carbon emission threshold as a high-emission time interval, and acquiring train operation data in the high-emission time interval to obtain a set of train operation status information;

[0010] A cluster analysis is performed on the set of train operation status information to determine the main influencing factors that lead to high carbon emissions of the train.

[0011] According to another aspect of the present application, a carbon emission monitoring system for green transportation in railway engineering is provided, comprising:

[0012] a time series data acquisition module, configured to obtain actual power consumption during train operation from the train traction power supply system based on a predetermined time period to obtain a time series of actual power consumption, and simultaneously obtain a power grid carbon emission factor corresponding to the train operation area from the power grid energy management system based on the predetermined time period to obtain a time series of power grid carbon emission factors;

[0013] a carbon emissions calculation module, configured to calculate a time series of train carbon emissions based on the time series of actual power consumption and the time series of the grid carbon emission factor;

[0014] a high-emission identification module, configured to mark data items in the time series of carbon emissions of the train that are greater than a preset carbon emission threshold as high-emission time intervals, and obtain train operation data in the high-emission time intervals to obtain a set of train operation status information;

[0015] The cluster analysis module is used to perform cluster analysis on the set of train operation status information to determine the main influencing factors that lead to high carbon emissions of the train.

[0016] The present application has at least the following technical effects: Compared with the existing technology, the carbon emission monitoring system and method for green transportation of railway engineering provided by the present application first obtains the actual power consumption of the train during operation from the train traction power supply system based on a predetermined time period, and simultaneously obtains the carbon emission factor of the power grid corresponding to the train operation area, and calculates the carbon emissions of the train in each time period based on this. Then, it further screens out high-carbon emission data items, retrieves the train operation status information in each high-carbon emission time period, and then, based on deep learning technology, performs deep feature extraction and cluster analysis on the train operation status information in each high-carbon emission time period to identify the main influencing factors that lead to high carbon emissions of the train. The present application can realize refined attribution analysis of train carbon emissions, timely discover and warn potential high-carbon emission events, and thus provide a scientific basis for formulating targeted energy conservation and emission reduction strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 Flowchart of a carbon emission monitoring method for green transportation in railway engineering according to an embodiment of the present application.

[0019] Figure 2 This is a flowchart of sub-step S4 of the carbon emission monitoring method for green transportation in railway engineering according to an embodiment of the present application.

[0020] Figure 3 Schematic diagram of data flow of sub-step S4 of the carbon emission monitoring method for green transportation of railway engineering according to an embodiment of the present application.

[0021] Figure 4 This is a flowchart of sub-step S42 of the carbon emission monitoring method for green transportation in railway engineering according to an embodiment of the present application.

[0022] Figure 5 This is a flowchart of sub-step S422 of the carbon emission monitoring method for green transportation in railway engineering according to an embodiment of the present application.

[0023] Figure 6 4 is a block diagram of a carbon emission monitoring system for green transportation in railway engineering according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0025] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0026] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0027] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0028] It should be noted in advance that the acquisition and processing of all information or data in this application are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization of the authority administrator.

[0029] Figure 1 Flowchart of the carbon emission monitoring method for green transportation of railway engineering according to the embodiment of the present application. Figure 1 As shown, the carbon emission monitoring method for green transportation of railway engineering includes the following steps: S1, obtaining the actual power consumption during the operation of the train from the train traction power supply system based on a predetermined time period to obtain a time series of the actual power consumption, and at the same time obtaining the grid carbon emission factor corresponding to the train operation area from the grid energy management system based on the predetermined time period to obtain a time series of the grid carbon emission factor; S2, calculating the time series of the train carbon emissions based on the time series of the actual power consumption and the time series of the grid carbon emission factor; S3, marking the data items in the time series of the train carbon emissions that are greater than a preset carbon emission threshold as high emission time intervals, and obtaining the train operation data in the high emission time interval to obtain a set of train operation status information; S4, performing cluster analysis on the set of train operation status information to determine the main influencing factors leading to high carbon emissions of the train.

[0030] In the aforementioned carbon emissions monitoring method for green transportation in railway projects, step S1 involves obtaining actual power consumption during train operation from the train's traction power supply system over a predetermined time period to obtain a time series of actual power consumption. Simultaneously, the grid carbon emission factor corresponding to the train's operating region is obtained from the grid energy management system over the predetermined time period to obtain a time series of the grid carbon emission factor. It should be understood that dynamic monitoring of railway carbon emissions requires capturing both the spatiotemporal variations in train energy consumption and grid carbon emission intensity. Traditional methods employ fixed carbon emission factors (e.g., national annual averages) and monthly energy consumption statistics, resulting in calculations that fail to reflect the true impact of the grid's real-time energy mix (e.g., intraday fluctuations in wind and photovoltaic power generation) on carbon emissions. Therefore, for trains operating across regions (e.g., passing through thermal power-dominated areas and hydropower-rich areas), this application simultaneously obtains traction power supply system power consumption and regional grid carbon emission factors to accurately calculate the train's carbon emission intensity in different operating sections, avoiding overestimation or underestimation of carbon emissions due to fixed factors, thereby enabling dynamic monitoring of train carbon emissions.

[0031] Specifically, because trains may cross multiple different grid coverage areas, the grid energy mix (e.g., the proportion of wind power and photovoltaic power generation) in each area may vary significantly. Therefore, the carbon emission factor used must be dynamically adjusted based on the train's actual location. Given that the grid energy mix fluctuates over time and with factors such as weather conditions, selecting an appropriate scheduled time period is particularly important. The diurnal power generation of some renewable energy sources (such as solar and wind) can fluctuate significantly due to weather conditions. This means that setting a scheduled time period that is too long may mask the impact of these short-term fluctuations on carbon emissions. Conversely, setting a short time period may result in excessive data processing and increase the computational burden. Choosing an appropriate scheduled time period is key to ensuring accurate calculation results. Typically, a suitable scheduled time period can be determined based on the average train speed and the frequency of changes in the grid energy mix, balancing computational accuracy and efficiency. For example, a 15-minute time period can capture short-term fluctuations in the grid energy mix without excessive data processing.

[0032] During implementation, an effective data verification mechanism must be established to ensure the accuracy of data collected from the train's traction power supply system and the grid energy management system. This mechanism should include, but is not limited to, data format consistency checks, data integrity verification, and the identification and handling of outliers. For example, when a data anomaly is detected during a specific time period, the system should automatically trigger an alarm and attempt to reacquire or repair the data for that period to ensure the continuity and integrity of the data link. Furthermore, consideration must be given to special circumstances during train operation, such as when a train traverses a section with a sudden gradient or undergoes rapid acceleration, where energy consumption patterns can change significantly. Data generated under these special circumstances is equally important for understanding the train's carbon emissions characteristics. Therefore, when designing a data collection plan, special attention should be paid to data collection methods under these special operating conditions to ensure that data from these critical moments is not missed.

[0033] Furthermore, the impact of different seasons and weather conditions on the grid's energy mix must be considered. For example, in hot summer weather, increased air conditioning loads can lead to higher grid loads, which in turn affects the grid's carbon emissions factor. Increased heating demand in winter can have a similar effect. Therefore, when setting a predetermined time period, the impact of seasonal factors on the grid's carbon emissions factor must also be considered. By introducing more refined time divisions and more flexible data collection strategies, we can ensure that the final calculation results truly reflect the carbon emissions during train operation.

[0034] To better address potential data loss or corruption, a robust data backup and recovery mechanism must be established. This includes not only designing data storage redundancy at the physical level but also applying data validation and error correction algorithms at the logical level. For example, advanced error-correcting coding techniques can repair damaged data to a certain extent, reducing analytical bias caused by data loss. Furthermore, regular data maintenance and cleanup, removing redundant and outdated data and keeping the database clean and efficient, are also essential for ensuring data quality.

[0035] In the above-mentioned carbon emission monitoring method for green transportation of railway engineering, the step S2 calculates the time series of the train's carbon emissions based on the time series of the actual power consumption and the time series of the grid carbon emission factor. It should be understood that traditional carbon emission accounting adopts the static mode of "total power consumption × fixed factor". Due to the differences in energy structures in different regions, the carbon emission intensity of a train passing through a hydropower-rich area at night may be only 30% of that passing through a coal-fired power grid area during the day. The present application obtains the carbon emissions of the train in each time period by multiplying the time series of the actual power consumption and the time series of the grid carbon emission factor point by point, and constructs a dynamic change curve of carbon emissions, which helps to accurately locate high-emission periods and sections.

[0036] Specifically, to calculate the time series of a train's carbon emissions, the time series of actual power consumption must be multiplied point by point by the time series of the grid's carbon emission factor. This process not only reflects the train's carbon emissions over different time periods but also reveals how carbon emission intensity varies over time and space, helping to accurately locate high-emission periods and sections. To achieve this, seamless data interfaces must be ensured, allowing the actual power consumption data of the train's traction power supply system to be matched with the carbon emission factor data from the grid energy management system on the same time basis. This requires addressing cross-system and cross-departmental data sharing and establishing efficient data transmission channels to ensure real-time data updates and synchronized processing. Given that trains may traverse multiple different grid coverage areas, each with significantly different grid energy mixes (e.g., the proportion of wind power and photovoltaic power generation), the carbon emission factor used must be dynamically adjusted based on the train's actual location during the data matching process. Only in this way can the calculated results truly reflect the carbon emission impact of the electricity consumed during train operation.

[0037] In the above-mentioned carbon emission monitoring method for green transportation in railway engineering, step S3 marks the data items in the time series of the train carbon emissions that are greater than a preset carbon emission threshold as high-emission time intervals, and obtains the train operation data in the high-emission time interval to obtain a collection of train operation status information. It should be understood that since the optimization of train carbon emissions needs to prioritize high-emission scenarios, this application further sets a preset carbon emission threshold to screen out high-carbon emission events during train operation, so as to quickly locate the train operation conditions that require key analysis, and extracts multi-dimensional train operation status information within each high-emission time interval for in-depth analysis, in order to find carbon emission optimization space and provide strong data support for the formulation of energy-saving and emission reduction measures. In a specific example of this application, the train operation status information includes ambient temperature, ambient wind speed, mileage, average train speed, track segment slope, track segment curvature, actual power consumption, grid carbon emission factor, and train carbon emissions. Among them, ambient temperature affects the heat dissipation efficiency and cooling / heating requirements of train equipment, thereby affecting energy consumption; ambient wind speed may affect the train's running resistance and aerodynamic performance; mileage and average train speed reflect the train's overall operating efficiency and energy consumption level; track section slope and curvature reflect the terrain undulations and turning changes during train operation, directly affecting the train's traction consumption and energy efficiency during operation; and actual power consumption directly reflects the train's energy consumption during operation; the grid carbon emission factor reflects the carbon emission intensity of the electricity used by the train, revealing the impact of the power grid's energy structure on carbon emissions; and train carbon emissions are the result of the combined effect of all the above factors, directly reflecting the level of carbon emissions during train operation. Through a comprehensive analysis of the above multi-dimensional information, it is helpful to identify the contribution of different influencing factors to train carbon emissions, thereby providing more refined guidance for the subsequent formulation of energy conservation and emission reduction strategies.

[0038] In the above-mentioned carbon emission monitoring method for green transportation of railway engineering, the step S4 performs cluster analysis on the set of train operation status information to determine the main influencing factors that lead to high carbon emissions of trains. Specifically, since high carbon emissions are usually the result of the coupling of multiple factors, traditional methods (such as linear regression) are difficult to analyze nonlinear correlations. For example, under the same slope, an increase in ambient wind speed may cause a nonlinear increase in traction energy consumption. Therefore, this application further introduces a deep learning algorithm to perform deep feature extraction and cluster analysis on the train operation status information in each high emission time interval, so as to mine potential correlation patterns from multi-source heterogeneous data and identify the dominant influencing factors. Among them, Figure 2 This is a flowchart of sub-step S4 of the carbon emission monitoring method for green transportation in railway engineering according to an embodiment of the present application. Figure 3Schematic diagram of data flow of sub-step S4 of the carbon emission monitoring method for green transportation of railway engineering according to the embodiment of the present application. Figure 2 and Figure 3 As shown, the step S4 includes the steps of: S41, performing structured embedding coding on each train operation status information in the set of train operation status information to obtain a set of train operation status feature embedding coding vectors; S42, performing train operation status feature gain aggregation based on self-learning reinforcement on the set of train operation status feature embedding coding vectors to obtain a train operation status feature aggregation analysis modeling vector; S43, performing a carbon emission influencing factor identification module based on the train operation status feature aggregation analysis modeling vector to obtain an identification result.

[0039] Specifically, in one specific example of the present application, step S41 includes inputting each piece of train operation status information from the set of train operation status information into an embedded coding module based on a multi-layer perceptron (MLP) to obtain a set of embedded coding vectors for the train operation status features. It should be understood that because the train operation status information contains multi-source heterogeneous data such as the environment, track, train operation conditions, and grid energy structure, it has high dimensionality and complex nonlinear coupling relationships between parameters (such as the nonlinear correlation between track segment slope and curvature and average train speed). Directly using the raw data for cluster analysis will lead to the "curse of dimensionality" and make it difficult to capture the deep interactive features between the multi-source parameters. Therefore, the present application further embeds and codes each piece of train operation status information using a multi-layer perceptron (MLP) model to map the high-dimensional, heterogeneous operation status information into a low-dimensional continuous vector space, eliminating redundant noise while preserving key features, and providing a unified representation for subsequent analysis. Specifically, first, the train operation status information (ambient temperature, wind speed, slope, speed, etc.) in each time window is normalized and spliced ​​into an input vector, which is then input into the multi-layer perceptron model. The multi-layer perceptron maps the input vector into a low-dimensional, continuous feature embedding coding vector through a combination of multiple layers of nonlinear transformations and activation functions, thereby obtaining a set of train operation status feature embedding coding vectors.

[0040] Specifically, step S42 performs train operation state feature gain aggregation based on self-learning reinforcement on the set of train operation state feature embedded coding vectors to obtain a train operation state feature aggregate analysis modeling vector. It should be understood that the set of train operation state feature embedded coding vectors reflects the multi-dimensional characteristics of the train operation state within each time window. Due to the time-varying nature of train operation conditions and external influencing factors, the feature distributions of each train operation state feature embedded coding vector vary to a certain extent. In order to further mine the key features that have a significant impact on the high carbon emissions of the train from the train running state feature embedding coding vectors in each high carbon emission time window, this application proposes a train running state feature gain aggregation method based on self-learning reinforcement, which performs cluster analysis on the set of train running state feature embedding coding vectors to mine the global dominant mode of the train running state, and uses the preliminary clustering results as the initial clustering center of self-supervised learning, and performs feature gain evaluation on each train running state feature embedding coding vector to determine its gain contribution to the train carbon emissions, thereby guiding the clustering center to perform adaptive adjustment in the direction of high gain features, thereby accurately capturing the key features of the train's high carbon emissions and obtaining the train running state feature aggregation analysis modeling vector. Among them, Figure 4 FIG4 is a flow chart of sub-step S42 of the carbon emission monitoring method for green transportation of railway engineering according to an embodiment of the present application. Figure 4 As shown, the step S42 includes the steps of: S421, performing linear clustering analysis on the set of the train running state feature embedded coding vectors to obtain the train running state feature initial linear clustering center coding vector; S422, calculating the linear clustering compensation component between the set of the train running state feature embedded coding vectors and the train running state feature initial linear clustering center coding vector to obtain the train running state feature linear clustering compensation component coding vector; S423, fusing the train running state feature linear clustering compensation component coding vector and the train running state feature initial linear clustering center coding vector to obtain the train running state feature aggregation analysis modeling vector.

[0041] More specifically, step S421 is expressed as follows:

[0042] ;

[0043] ;

[0044] in, represents the set of embedding encoding vectors of train running status features, 、 、 and They represent the first, second, and third vectors in the set of train running status feature embedding coding vectors. and The train running status feature embedding encoding vector, represents the number of train running status feature embedding coding vectors in the set of train running status feature embedding coding vectors, Represents the initial linear cluster center encoding vector of the train running status characteristics.

[0045] That is, by leveraging the global feature skeleton of linear clustering, the high-dimensional train operation status information is projected into a low-dimensional subspace defined by the cluster centers. By extracting the main linear components, the global distribution patterns of the main energy consumption patterns are retained, and high-frequency noise interference is filtered out through low-rank approximation, forming a physically meaningful benchmark feature framework. Through this linear clustering approach, a dual-separation feature expression mechanism is constructed. The generated initial linear cluster center encoding vector of the train operation status feature serves as an explicit representation of the linear structure, which not only frames the macro-trend boundary of carbon emission changes, but also compresses nonlinear coupling factors into the residual space, allowing subsequent deep networks to focus on exploring deep nonlinear attribution paths, thereby improving the accuracy of tracing the source of sudden high-emission events and the effectiveness of regulation.

[0046] Figure 5 FIG4 is a flow chart of sub-step S422 of the carbon emission monitoring method for green transportation of railway engineering according to an embodiment of the present application. Figure 5 As shown, the step S422 includes the steps of: S4221, constructing a deep collaborative implicit coding vector between each train running state feature embedded coding vector in the set of the train running state feature embedded coding vector and the train running state feature initial linear clustering center coding vector to obtain a set of train running state feature deep collaborative implicit coding vectors; S4222, based on the set of the train running state feature deep collaborative implicit coding vectors, calculating the clustering compensation gain factor of each train running state feature embedded coding vector in the set of the train running state feature embedded coding vector relative to the train running state feature initial linear clustering center coding vector to obtain a set of train running state clustering compensation gain factors; S4223, based on the set of the train running state clustering compensation gain factors, calculating the linear clustering compensation component coding vector of the set of the train running state feature embedded coding vector to obtain the train running state feature linear clustering compensation component coding vector.

[0047] In a specific example of the present application, step S4221 includes: first, performing activation processing based on the Sigmoid function on the initial linear clustering center encoding vector of the train running state feature to obtain the initial linear clustering center activation encoding vector of the train running state feature, which is expressed as follows:

[0048] ;

[0049] in, represents the Sigmoid function, Represents the initial linear cluster center activation encoding vector of the train running status features.

[0050] That is, the linear coding value is compressed to the (0,1) interval through the Sigmoid activation function, which can not only suppress the interference of extreme values, but also convert the physical quantity into a probabilistic representation, so that the generated initial linear clustering center activation coding vector of the train operation status characteristics not only retains the global correlation characteristics of key parameters, but also enhances the nonlinear sensitivity to secondary factors, thereby providing more discriminative feature input for subsequent network layers.

[0051] Then, the initial linear cluster center activation coding vector of the train running state feature is cascaded and fused with each train running state feature embedding coding vector in the set of the train running state feature embedding coding vector, and then input into the deep collaborative feature extraction network to obtain the set of deep collaborative implicit coding vectors of the train running state feature, which can be expressed as follows:

[0052] ;

[0053] in, represents the bias term, represents the weight matrix, Represents a cascade function, The first vector in the set of deep collaborative implicit coding vectors representing train running status features Deep collaborative implicit encoding vector of train running status features.

[0054] That is, through the cascade fusion operation, the initial linear clustering center activation coding vector of the train running status characteristics and the train running status feature embedding coding vector are cross-level feature splicing, which not only retains the framework constraint of linear clustering on the benchmark emission pattern, but also injects the original fine-grained dynamic details. Then, through the deep learning ability of the neural network, the deep nonlinear collaborative features of the original train running status feature embedding coding vector relative to the initial linear clustering center are automatically mined to form a deep collaborative implicit coding vector of the train running status characteristics with both global framework constraints and local detail sensitivity, providing a refined feature basis with both interpretability and discriminability for the subsequent feature compensation gain calculation.

[0055] In a specific example of the present application, step S4222 is expressed as follows:

[0056] ;

[0057] in, represents the logarithmic function with base 2, express Middle The eigenvalues ​​at the positions, express Middle The eigenvalues ​​at the positions, Represents the characteristic scale of the deep collaborative implicit coding vector of the train running status feature, express The corresponding train operation status clustering compensation gain factor, Represents an exponential function with base e.

[0058] That is, by introducing the calculation of the clustering compensation gain factor, the high-order feature information in the implicit coding vector of the train operation status characteristics is deeply coordinated to utilize the differentiated performance of the carbon emission characteristics under different operation scenarios relative to the linear clustering benchmark, so as to realize the fine adjustment of the initial linear clustering structure based on the obtained train operation status clustering compensation gain factor, form a global nonlinear compensation strategy, and make the final clustering result more in line with the actual data distribution, effectively distinguish the key driving factors of high carbon emissions, and provide causal decision support for the formulation of targeted energy-saving and emission reduction measures, while enhancing the generalization ability and robustness of the model for complex scenarios.

[0059] In particular, in a preferred example of the present application, the step S4223 includes: first, based on the synergy strength between the train running state feature deep collaborative implicit coding vector and the train running state feature embedded coding vector corresponding to each group in the set of the train running state feature deep collaborative implicit coding vector and the set of the train running state feature embedded coding vector, each train running state clustering compensation gain factor in the set of the train running state clustering compensation gain factor is subjected to clustering balance correction to obtain a set of corrected train running state clustering compensation gain factors, which is expressed as follows:

[0060] ;

[0061] ;

[0062] ;

[0063] in, Indicates the calculation of inner product, express and The clustering active items between It means calculating the variance of the set of all eigenvalues ​​in the vector. Representing and clustering active items The relevant divergence factor, for The clustering delay of For the scattering relaxation of linear cluster centers, represents an exponential function with a natural constant as the base, represents the cluster compensation gain factor of the corrected train operation status, is the absolute value function.

[0064] Among them, considering that when calculating the train running state clustering compensation gain factor, the incremental nonlinear information of the linear clustering result based on the train running state feature deep collaborative implicit coding vector is added in addition to the original train running state feature embedded coding vector, resulting in the addition of cluster elements in the cluster space, thereby causing the system non-equilibrium state of the cluster space distribution. In this regard, the present application quantifies the synergy strength between the train running state feature deep collaborative implicit coding vector and the train running state feature embedded coding vector by constructing a clustering active term, and by establishing constraints under the fractal structure, the train running state clustering compensation gain factor is balanced and corrected, so that the train running state clustering compensation gain factor can adaptively adjust the resonance amplitude during the propagation process of the fractal space, thereby suppressing the interference of abnormal energy fluctuations on the clustering stability, ensuring that the corrected compensation operator retains the nonlinear correction advantage and maintains the global energy balance of the clustering system, and realizes the clustering equilibrium state optimization under the fractal structure. In this way, clustering incremental compensation based on the corrected train running state clustering compensation gain factor can make the clustering result have stronger noise resistance and working condition adaptability while maintaining nonlinear accuracy.

[0065] Then, the set of the corrected train running state cluster compensation gain factors is normalized based on the Softmax function to obtain a set of normalized train running state cluster compensation gain factors, which is expressed as follows:

[0066] ;

[0067] in, represents the softmax function, express The corresponding normalized train operation status clustering compensation gain factor.

[0068] That is, by using the Softmax function to convert the corrected train operation state clustering compensation gain factor into a standardized parameter within the probability domain, not only does it eliminate the dimensional differences between different feature dimensions, but it also strengthens the dominant role of the key compensation factor through the exponential decay mechanism, while suppressing minor noise interference. This allows for dynamic balance adjustment of the compensation increment based on the generated normalized train operation state clustering compensation gain factor through probability space mapping. It should be understood that the Softmax function processing ensures that the probability value of each corrected train operation state clustering compensation gain factor is positively correlated with its actual impact intensity. This results in a weight distribution result that conforms to cognitive logic while maintaining the advantages of nonlinear correction, resolves the numerical instability problem of the original operator set, and enhances the model's anti-interference ability through the global normalization characteristics of the probability distribution.

[0069] Finally, based on the set of normalized train running state clustering compensation gain factors, the set of train running state feature embedded coding vectors is weightedly aggregated to obtain the train running state feature linear clustering compensation component coding vector, which is expressed as follows:

[0070] ;

[0071] in, Represents the linear clustering compensation component encoding vector of the train running status characteristics.

[0072] That is, through weighted aggregation, the local optimization of the normalized train operation status feature clustering compensation increment is transformed into a systematic improvement of the overall clustering structure. Specifically, the weight allocation mechanism based on the normalized train operation status clustering compensation gain factor can quantify the differences in the contribution of different operation status features to the cluster center correction, so that the weighted train operation status feature embedding coding vector not only retains the key information of the original feature distribution, but also integrates the global adjustment effect of nonlinear compensation. Through this dynamic weighting method, a precise compensation mapping for linear clustering deviation is formed, so that the generated train operation status feature linear clustering compensation component coding vector can effectively reconcile the energy distribution contradiction between the original feature space and the cluster compensation space, and improve the stability and accuracy of the train operation status feature clustering results.

[0073] More specifically, step S423 is expressed as follows:

[0074] ;

[0075] in, and Represents different weight parameters, Represents the train running status feature aggregation analysis modeling vector.

[0076] Specifically, by organically integrating the global structural benchmark provided by the initial linear cluster center encoding vector of the train operating state characteristics with the nonlinear detail corrections carried by the linear cluster compensation component encoding vector of the train operating state characteristics, a composite feature representation is formed that accounts for both macroscopic regularities and microscopic fluctuations. Through this fusion approach, the generated train operating state characteristic aggregation analysis modeling vector becomes a bridge between global statistical characteristics and local anomaly detection, providing a core parameter carrier that combines physical interpretability and mathematical rigor for subsequent carbon emission attribution analysis.

[0077] Specifically, in a specific example of the present application, step S43 includes: inputting the train operation status feature aggregation analysis modeling vector into a classifier-based carbon emission influencing factor identification module to obtain an identification result, wherein the identification result includes the main influencing factors that lead to high carbon emissions of the train and their corresponding confidence levels, wherein the main influencing factors include environmental condition influencing factors, track topography influencing factors, train operation condition influencing factors, and power grid energy structure influencing factors. Specifically, during the training process, the classifier-based carbon emission influencing factor identification module uses the feature data of historical high carbon emission events and their corresponding influencing factor labels to perform supervised learning, and continuously optimizes model parameters to improve recognition accuracy. During the recognition process, the classifier receives the train operation status feature aggregation analysis modeling vector as input, performs layer-by-layer feature analysis on it based on the learned classification decision boundary, and finally calculates the probability distribution of each influencing factor through the Softmax function of the output layer, that is, the main influencing factors that lead to high carbon emissions of the train and their corresponding confidence levels, so as to provide an intuitive basis for energy conservation and emission reduction for train operation management. For example, when the identification results show that environmental conditions (such as ambient temperature and wind speed) are the main factors leading to high carbon emissions from trains, measures such as adjusting train running times, strengthening train insulation or ventilation can be considered to reduce energy consumption; when track terrain factors (such as slope and curvature) dominate, then consideration can be given to optimizing line design or adopting more energy-efficient train traction technology; when train operating conditions (such as train speed and acceleration control) are key, carbon emissions can be reduced by optimizing train operation strategies and improving driving efficiency; and when the power grid energy structure factors (such as carbon emission factors) are more significant, consideration can be given to optimizing train scheduling strategies or promoting the transformation of the power grid energy structure to reduce carbon emissions.

[0078] In summary, a carbon emission monitoring method for green transportation in railway engineering based on the embodiment of the present application is explained. It first obtains the actual power consumption of the train during operation from the train traction power supply system based on a predetermined time period, and simultaneously obtains the grid carbon emission factor corresponding to the train operation area. Based on this, the carbon emissions of the train in each time period are calculated. Then, high-carbon emission data items are further screened out, and the train operation status information in each high-carbon emission time period is retrieved. Then, based on deep learning technology, deep feature extraction and cluster analysis are performed on the train operation status information in each high-carbon emission time period to identify the main influencing factors that lead to high carbon emissions of the train. In this way, it is possible to achieve refined attribution analysis of train carbon emissions, timely discover and warn potential high-carbon emission events, and thus provide a scientific basis for formulating targeted energy-saving and emission reduction strategies.

[0079] Furthermore, a carbon emission monitoring system for green transportation in railway engineering is also provided.

[0080] Figure 6 FIG is a block diagram of a carbon emission monitoring system for green transportation in railway engineering according to an embodiment of the present application. Figure 6 As shown, the carbon emission monitoring system 100 for green transportation of railway engineering according to an embodiment of the present application includes: a time series data acquisition module 110, which is used to obtain the actual power consumption of the train during operation from the train traction power supply system based on a predetermined time period to obtain a time series of the actual power consumption, and at the same time obtain the grid carbon emission factor corresponding to the train operation area from the grid energy management system based on the predetermined time period to obtain a time series of the grid carbon emission factor; a carbon emission calculation module 120, which is used to calculate the time series of the train carbon emissions based on the time series of the actual power consumption and the time series of the grid carbon emission factor; a high emission identification module 130, which is used to mark the data items in the time series of the train carbon emissions that are greater than a preset carbon emission threshold as high emission time intervals, and obtain the train operation data in the high emission time interval to obtain a set of train operation status information; a cluster analysis module 140, which is used to perform cluster analysis on the set of train operation status information to determine the main influencing factors leading to high carbon emissions of the train.

[0081] The basic principles of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in the present invention are merely illustrative and non-limiting, and should not be construed as necessarily possessed by each embodiment of the present invention. Furthermore, the specific details of the above embodiments are provided for illustrative purposes and to facilitate understanding, and are not intended to be limiting. These details do not necessarily limit the present invention to being implemented using these specific details.

[0082] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0083] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be encompassed therein. Any reference to a figure in a claim should not be construed as limiting the claim to which it relates.

[0084] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0085] Finally, it should be noted that the above description has been provided for purposes of illustration and description. Furthermore, the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to be limiting. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art will appreciate that the technical solutions of the present invention may be modified or replaced with equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A carbon emission monitoring method for green transportation in railway engineering, characterized in that: include: Acquire actual power consumption during train operation from the train traction power supply system based on a predetermined time period to obtain a time series of actual power consumption, and simultaneously acquire a power grid carbon emission factor corresponding to the train operation area from the power grid energy management system based on a predetermined time period to obtain a time series of the power grid carbon emission factor; Calculate the time series of train carbon emissions based on the time series of actual power consumption and the time series of grid carbon emission factors; Mark the data items in the time series of train carbon emissions that are greater than a preset carbon emission threshold as high-emission time intervals, and obtain the train operation data in the high-emission time interval to obtain a set of train operation status information; Cluster analysis is performed on the collection of train operation status information to identify the main factors leading to high carbon emissions from trains, including: Performing structured embedding coding on each piece of train running status information in the set of train running status information to obtain a set of train running status feature embedding coding vectors; Performing linear clustering analysis on the set of train running state feature embedding coding vectors to obtain the initial linear clustering center coding vector of the train running state feature; Constructing a deep collaborative implicit coding vector between each train running state feature embedding coding vector in the set of train running state feature embedding coding vectors and the train running state feature initial linear clustering center coding vector to obtain a set of train running state feature deep collaborative implicit coding vectors; Based on the set of train running state feature deep collaborative implicit coding vectors, calculating the clustering compensation gain factor of each train running state feature embedded coding vector in the set of train running state feature embedded coding vectors relative to the initial linear clustering center coding vector of the train running state feature, to obtain a set of train running state clustering compensation gain factors; Based on the set of train running state clustering compensation gain factors, calculating the linear clustering compensation component code vector of the set of train running state feature embedding code vectors to obtain the train running state feature linear clustering compensation component code vector; Fusing the linear clustering compensation component encoding vector of the train running state feature and the initial linear clustering center encoding vector of the train running state feature to obtain the train running state feature aggregation analysis modeling vector; A carbon emission influencing factor identification module is performed based on the train operation status feature aggregation analysis modeling vector to obtain the identification results.

2. The carbon emission monitoring method for green transportation of railway engineering according to claim 1, characterized in that: Train operation status information includes ambient temperature, ambient wind speed, operating mileage, average train speed, track section slope, track section curvature, actual power consumption, grid carbon emission factor and train carbon emissions.

3. The carbon emission monitoring method for green transportation in railway engineering according to claim 2 is characterized in that: Perform structured embedding coding on each piece of train running status information in the set of train running status information to obtain a set of train running status feature embedding coding vectors, including: Each piece of train running status information in the set of train running status information is input into an embedding coding module based on a multi-layer perceptron to obtain a set of train running status feature embedding coding vectors.

4. The carbon emission monitoring method for green transportation in railway engineering according to claim 3 is characterized in that: Constructing a deep collaborative implicit coding vector between each train running state feature embedding coding vector in the set of train running state feature embedding coding vectors and the train running state feature initial linear clustering center coding vector to obtain a set of train running state feature deep collaborative implicit coding vectors, including: Performing activation processing based on the Sigmoid function on the initial linear cluster center encoding vector of the train running state feature to obtain the initial linear cluster center activation encoding vector of the train running state feature; The initial linear clustering center activation coding vector of the train running state feature is cascaded and fused with each train running state feature embedding coding vector in the set of train running state feature embedding coding vectors, and then input into the deep collaborative feature extraction network to obtain a set of deep collaborative implicit coding vectors of the train running state feature.

5. The carbon emission monitoring method for green transportation in railway engineering according to claim 4 is characterized in that: Based on the set of train running state clustering compensation gain factors, a linear clustering compensation component encoding vector of the set of train running state feature embedding encoding vectors is calculated to obtain a train running state feature linear clustering compensation component encoding vector, including: Based on the synergistic effect strength between each corresponding group of train running state feature deep collaborative implicit coding vectors and train running state feature embedded coding vectors in the set of train running state feature deep collaborative implicit coding vectors and the set of train running state feature embedded coding vectors, cluster balance correction is performed on each train running state cluster compensation gain factor in the set of train running state cluster compensation gain factors to obtain a set of corrected train running state cluster compensation gain factors; performing a normalization process based on a Softmax function on the set of corrected train running state clustering compensation gain factors to obtain a set of normalized train running state clustering compensation gain factors; Based on a set of normalized train running state clustering compensation gain factors, a set of train running state feature embedding coding vectors is weightedly aggregated to obtain a train running state feature linear clustering compensation component coding vector.

6. The carbon emission monitoring method for green transportation in railway engineering according to claim 5, characterized in that: Based on the train operation status feature aggregation analysis modeling vector, the carbon emission influencing factor identification module is used to obtain the identification results, including: The train operation status feature aggregation analysis modeling vector is input into the classifier-based carbon emission influencing factor identification module to obtain the identification results. The identification results include the main influencing factors that lead to high carbon emissions of trains and their corresponding confidence levels. Among them, the main influencing factors include environmental conditions, track terrain, train operation conditions, and grid energy structure.

7. A carbon emission monitoring system for green transportation in railway engineering, used to execute the method according to any one of claims 1 to 6, characterized in that: include: a time series data acquisition module for acquiring actual power consumption during train operation from the train traction power supply system based on a predetermined time period to obtain a time series of actual power consumption, and simultaneously acquiring a power grid carbon emission factor corresponding to the train operation area from the power grid energy management system based on a predetermined time period to obtain a time series of the power grid carbon emission factor; A carbon emissions calculation module is used to calculate the time series of train carbon emissions based on the time series of actual power consumption and the time series of grid carbon emission factors; A high-emission identification module is used to mark data items in the time series of train carbon emissions that are greater than a preset carbon emission threshold as high-emission time intervals, and obtain train operation data in the high-emission time interval to obtain a collection of train operation status information; The cluster analysis module is used to perform cluster analysis on a collection of train operation status information to determine the main influencing factors that lead to high carbon emissions of trains.

Images