Railway traffic carbon footprint dynamic tracking system and method based on block chain and edge calculation

Abstract

The invention relates to the technical field of carbon footprint dynamics, in particular to a railway traffic carbon footprint dynamic tracking system and method based on a block chain and edge calculation. The data acquisition unit is used for acquiring train operation state, environment parameters and traction energy consumption data; the edge calculation unit fuses the multi-source data acquired by the data acquisition unit through adaptive Kalman filtering to obtain standardized data, converts a train operation state into carbon emission based on an instantaneous power model, and introduces a dynamic emission factor to calculate and obtain accumulated carbon emission; and the block chain network unit is used for compressing the accumulated carbon emission data and then generating a Merkle tree root hash upper chain, and storing the carbon emission data to a block chain. Incremental uplink and rapid verification of the data are realized, the on-chain data processing efficiency and the query traceability are improved, and the problems of high storage pressure and low verification efficiency of a traditional on-chain evidence storage mode in a high-frequency track data writing scene are solved.

Description

Technical Field

[0001] The present invention relates to the field of dynamic carbon footprint technology, and more specifically, to a railway transportation carbon footprint dynamic tracking system and method based on blockchain and edge computing. Background Art

[0002] Emissions data statistics often rely on periodic summaries and macro-estimates, resulting in low data granularity and poor timeliness. Dynamic tracking can monitor emissions within seconds or minutes, enabling operators to promptly understand emission characteristics across different lines and operating conditions, facilitating targeted optimization and adjustments. However, traditional static rules do not respond promptly to sudden high-carbon emission events and have a high false alarm rate. Furthermore, traditional on-chain evidence storage methods place significant storage pressure in scenarios with high-frequency track data writes. Therefore, a system and method for dynamic tracking of railway transportation carbon footprints based on blockchain and edge computing is provided. Summary of the Invention

[0003] The purpose of the present invention is to provide a dynamic tracking system and method for the carbon footprint of railway transportation based on blockchain and edge computing, so as to solve the problems of untimely response and high false alarm rate to sudden high-carbon emission events under the traditional static rules proposed in the above background technology; and the high storage pressure of the traditional on-chain evidence storage method in the high-frequency track data writing scenario.

[0004] To achieve the above objectives, on the one hand, the present invention aims to provide a railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing, comprising: A data acquisition unit, which is used to collect train operation status, environmental parameters and traction energy consumption data; The edge computing unit uses adaptive Kalman filtering to fuse the multi-source data collected by the data acquisition unit to obtain standardized data, and converts the train operation status into carbon emissions based on the instantaneous power model, and introduces a dynamic emission factor Calculate cumulative carbon emissions ; A blockchain network unit, which is used to compress the accumulated carbon emission data, generate a Merkle tree root hash, and store the carbon emission data on the blockchain; Dynamic tracking unit, which corrects on-chain data and real-time edge computing data based on the LSTM prediction model, uses the NSGA-II multi-objective genetic algorithm with elite retention strategy to perform global optimization and generate Pareto solution sets , and the Pareto solution set Sent to the edge computing unit for closed-loop optimization.

[0005] As a further improvement of this technical solution, the data acquisition unit includes an on-board sensor module, a track environment monitoring module and an energy consumption metering module; Among them, the on-board sensor module is used to monitor the train's energy consumption, load, speed and location data; The track environment monitoring module is used to collect track slope, weather conditions, and signal system status parameters; The energy consumption metering module is used to collect energy consumption data of traction power supply systems and stations.

[0006] As a further improvement of this technical solution, the edge computing unit includes a local preprocessing node module, a distributed computing module and an edge prediction module; The local pre-processing node module uses adaptive Kalman filtering to fuse the multi-source data collected by the data acquisition unit to obtain pre-processed standardized data; The distributed computing module converts the train operation status into carbon emissions using an instantaneous power model based on the pre-processed standardized data; Among them, the train power at time is calculated based on the instantaneous power model. The instantaneous total power and increase the efficiency of the traction system Introduced instantaneous total power Get cumulative energy consumption , and based on real-time emission factors The accumulated energy consumption Converted into cumulative carbon emissions; The instantaneous power model is formed by coupling air resistance power, slope resistance power and inertia power; The edge prediction module predicts the carbon emission value based on the carbon emission prediction model And output Boolean exception flag by introducing an auxiliary branch for exception detection .

[0007] As a further improvement of this technical solution, the edge prediction module involves the following specific steps: A carbon emission prediction model is constructed using a long short-term memory network and an attention mechanism prediction model to predict carbon emission levels in the future. The preprocessed standardized data is used as the input of the carbon emission prediction model, and the predicted carbon emission value is output. ; An anomaly detection auxiliary branch constructed using binary classification logic is introduced into the carbon emission prediction model to output a Boolean anomaly flag. and prediction confidence values ; Finally, the carbon emission prediction model packages the regression results and classification results into a triplet .

[0008] As a further improvement of this technical solution, the blockchain network unit includes a blockchain storage module and a blockchain early warning module; The blockchain storage module is used to hash and compress the carbon emission data from the edge computing unit according to the time window, construct a dynamic shard Merkle tree, and write it into the blockchain to ensure that the data is tamper-proof and traceable; The Merkle tree adopts a dynamic sharding structure, and the sharding logic is based on the train operation interval. The blockchain early warning module determines the threshold value based on the carbon emission data stored on the chain and triggers the early warning event based on the threshold judgment algorithm, providing timely response for operation and maintenance and scheduling.

[0009] As a further improvement of this technical solution, the dynamic slicing rules involved in the train running section are as follows: The railway network consists of The tracks between adjacent stations constitute a basic operating range. : ; in, Indicates the starting station of the interval; Indicates the terminal station of the section; The minimum geographic coordinate boundary of the interval; Indicates the maximum geographic coordinate boundary of the interval; Each interval Corresponding to a dynamic shard : ; Where, Represents the shard key, which is determined by the running interval Location and time window Sure; Represents the carbon emission dataset stored in the shard; Indicates the initial capacity threshold of the shard; Indicates the shard split threshold; Indicates the shard merging threshold; Based on train position and time window Determine the shard key ; When the train enters a new section When the shard is created, the real-time location and timestamp are continuously checked to see if they meet the new shard creation logic. If they do, a new shard is triggered. create; When sharding The number of data records stored in exceeds the split threshold When , the split operation is triggered; When adjacent shards and When low load and physical proximity are met at the same time, the merge operation is triggered and adjacent shards and Merge into .

[0010] As a further improvement of this technical solution, the specific steps involved in constructing the Merkle tree are: For each shard Carbon emissions data block in , calculate the leaf node hash ; Hash all leaf nodes in the shard Sort by time and generate parent node hashes in pairs. Repeat the above steps until a unique shard root hash is generated. ; Hash all shard roots As a leaf node, build a global Merkle tree.

[0011] As a further improvement of this technical solution, in the blockchain early warning module, the specific steps involved in the threshold judgment algorithm judging the threshold and triggering the early warning event are: Based on the blockchain early warning module monitoring chain newly added shard root hash ; Locate the carbon emission dataset within the shard through on-chain pointer index and Merkle path ; Iterate over the carbon emissions dataset For each data, if the actual carbon emissions Exceeding the set threshold is recorded as an abnormal event ; At the same time, when new abnormal events are to be evaluated, the most recent On-chain at this moment sequence; Based on the threshold judgment algorithm and historical on-chain data, determine whether there is an abnormality: Where, Indicates the Abnormal carbon emission events detected at any time; Indicates the The actual carbon emission value detected at any moment; Indicates the current train running section Historical average carbon emissions; Indicates at time The on-chain continuous anomaly probability coefficient is a weighted indicator of the occurrence of continuous anomalies in the historical period, reflecting whether the current anomaly is in a period of continuous anomaly (a value close to 1 is more likely to be a continuous anomaly); Indicates The isolated anomaly penalty coefficient at the moment is used to suppress the impact of accidental or sporadic mutation events; Represents the weight coefficient of the historical deviation term, which is used to measure the impact intensity of the difference between the current emission value and the historical mean; represents the continuous abnormality discrimination weight coefficient; represents the isolated anomaly discrimination weight coefficient, which is used to control the influence of sudden outliers on the judgment results; and ; like Exceeding the preset alarm threshold , the exception enters the summary data structure ; Among them, abnormal events are confirmed and encapsulated into a summary data structure as off-chain warning records. ; The summary data structure The data is uploaded to the chain and sent synchronously to the dispatch center to trigger an operation and maintenance response.

[0012] As a further improvement of this technical solution, the dynamic tracking unit includes a collaborative correction module, a multi-objective optimization decision module, a full life cycle carbon integration module and a closed-loop strategy issuance module; Among them, the collaborative correction module is based on the summary data structure , triples and historical carbon emission series, and the LSTM prediction model is used to predict the efficiency of the traction system. and real-time emission factors Make short-term forecasts and corrections; The multi-objective optimization decision module takes edge computing delay, energy consumption, and uplink overhead as the objective function, and comprehensively considers the sampling interval , edge computing quota and chain frequency , using NSGA-II multi-objective genetic algorithm with elite retention strategy to perform global optimization and generate Pareto solution set ; The full life cycle carbon integration module calculates comprehensive carbon emissions based on the carbon footprint expansion model If comprehensive carbon emissions are detected When the preset threshold is exceeded, the on-chain carbon transaction is triggered, where the carbon footprint expansion model is based on the cumulative carbon emissions , it is constructed by introducing the indirect carbon emission factors generated during the use of infrastructure; The closed-loop strategy issuing module is used to set the Pareto solution Sent to the edge computing unit.

[0013] In another aspect, the present invention provides a method for dynamically tracking the carbon footprint of railway transportation based on blockchain and edge computing, which is used in any of the above-mentioned systems for dynamically tracking the carbon footprint of railway transportation based on blockchain and edge computing, comprising the following steps: S10.1. The edge computing unit integrates and processes the multi-source data collected by the data acquisition unit, calculates the train's real-time energy consumption using an instantaneous power model, introduces a dynamic emission factor, and converts it into cumulative carbon emissions. S10.2. The blockchain network unit compresses and encodes data from edge computing, constructs a hash structure of carbon emission data blocks through a dynamic sharding Merkle tree, and uploads it to the blockchain, enabling traceable evidence storage. When a train crosses an operating range or time window, it automatically triggers the creation, splitting, or merging of new shards. At the same time, by monitoring the newly added shard hash on the chain, based on the threshold judgment algorithm combined with historical data to identify abnormal events, and then upload the abnormal summary data structure to the chain and push it to the dispatch center to trigger the operation and maintenance response; S10.3. The on-chain data and edge prediction results are collaboratively corrected through the dynamic tracking unit, and the NSGA-II multi-objective optimization algorithm is used to generate the Pareto solution set. After comprehensively considering latency, energy consumption, and on-chain overhead, the optimal strategy is fed back to the edge computing unit from the closed-loop strategy distribution module to achieve full-process optimization.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. A railway transportation carbon footprint dynamic tracking system and method based on blockchain and edge computing. A dynamic sharded Merkle tree is constructed based on blockchain. Sharded hashing is performed on carbon emission data in different operating intervals to achieve cross-interval traceability and historical auditing. At the same time, shards are constructed according to train operation intervals and time windows, supporting splitting, merging and archiving. Based on the dynamic sharding mechanism of train operation intervals and time windows, the carbon emission data is sharded and hashed for notarization in combination with the Merkle tree structure, realizing incremental data on-chain and rapid verification, improving on-chain data processing efficiency and query traceability capabilities, and solving the problems of high storage pressure and low verification efficiency of traditional on-chain notarization methods in high-frequency track data writing scenarios.

[0015] 2. A railway transportation carbon footprint dynamic tracking system and method based on blockchain and edge computing builds a carbon emission prediction model based on LSTM and attention mechanism, integrates historical operation trajectories and introduces an auxiliary binary classification detection branch. The dynamic threshold is controlled by prediction confidence, which improves the recognition sensitivity and judgment accuracy of abnormal events, solving the problem of untimely response and high false alarm rate to sudden high-carbon emission events under traditional static rules.

[0016] 3. A railway transportation carbon footprint dynamic tracking system and method based on blockchain and edge computing uses the NSGA-II multi-objective optimization algorithm with an elite retention strategy to perform global optimization among sampling intervals, edge computing delays, and on-chain overheads, generate a Pareto solution set, and dynamically issue a scheduling strategy based on resource status. This significantly enhances the system's adaptability to resource fluctuations and network congestion, and solves the problems of unbalanced resource allocation and high response delay under a single-objective scheduling mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of the overall process of the present invention.

[0018] The meaning of each number in the figure is: 1. Data acquisition unit; 2. Edge computing unit; 21. Local preprocessing node module; 22. Distributed computing module; 23. Edge prediction module; 3. Blockchain network unit; 31. Blockchain storage module; 32. Blockchain early warning module; 4. Dynamic tracking unit; 41. Collaborative correction module; 42. Multi-objective optimization decision module; 43. Full life cycle carbon integration module; 44. Closed-loop strategy issuance module. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] Example 1: Please refer to Figure 1 As shown, a railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing is provided, including a data acquisition unit 1, which is used to collect train operation status, environmental parameters and traction energy consumption data.

[0021] The data acquisition unit 1 includes a vehicle-mounted sensor module, a track environment monitoring module and an energy consumption metering module.

[0022] Among them, the on-board sensor module is used to monitor the train's energy consumption, load, speed and location data.

[0023] The track environment monitoring module is used to collect track slope, weather conditions, and signal system status parameters.

[0024] The energy consumption metering module is used to collect energy consumption data of traction power supply systems and stations.

[0025] In this embodiment, the railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing also includes an edge computing unit 2, which fuses the multi-source data collected by the data acquisition unit 1 through adaptive Kalman filtering to obtain standardized data, and converts the train operation status into carbon emissions based on the instantaneous power model, and introduces a dynamic emission factor Calculate cumulative carbon emissions .

[0026] Among them, the dynamic emission factor model is constructed based on the LSTM time series prediction algorithm, and the construction process involved is: The historical multi-source standardized data obtained by preprocessing in the edge computing unit is organized into fixed-length time series windows as the model input sequence; The LSTM layer is followed by a fully connected regression output layer to predict the emission factor at the next moment; The network loss function uses weighted mean absolute error to improve the prediction stability near outliers; In the prediction phase, the model is based on the latest The characteristic sequence of time steps is input and the predicted emission factor at the current moment is output; The predicted value is then corrected using carbon emission observation data and energy consumption data within a historical sliding window to obtain a calibrated emission factor. This dynamic emission factor model can continuously predict changes in the grid's carbon emission intensity online during train operation, and calculate dynamic carbon emissions based on traction energy consumption. It is suitable for scenarios such as carbon footprint calculation, early warning judgment, and carbon trading response.

[0027] Specifically, the LSTM model is used to predict real-time emission factors, and a historical sliding window is introduced for error backtracking and deviation self-calibration. The sliding variance analysis of traction efficiency is combined to achieve regional carbon intensity adaptation, effectively suppressing the deviation in carbon emission estimation caused by regional power grid fluctuations or data drift, and solving the problem that fixed emission factors are difficult to adapt to different lines and time periods.

[0028] In this embodiment, the edge computing unit 2 includes a local preprocessing node module 21 , a distributed computing module 22 and an edge prediction module 23 .

[0029] The local pre-processing node module 21 uses an adaptive Kalman filter to fuse the multi-source data collected by the data collection unit 1 to obtain pre-processed standardized data.

[0030] The local pre-processing node module 21 is deployed on the edge server of the station to perform data cleaning, anomaly detection, and preliminary carbon emission calculation.

[0031] All parameters are aligned according to a unified timestamp (GNSS clock) with millisecond accuracy.

[0032] The distributed computing module 22 converts the train operation status into carbon emissions using an instantaneous power model based on the pre-processed standardized data.

[0033] Among them, the train power at time is calculated based on the instantaneous power model. The instantaneous total power and increase the efficiency of the traction system Introduced instantaneous total power Get cumulative energy consumption , and based on real-time emission factors The accumulated energy consumption Converted into cumulative carbon emissions.

[0034] The instantaneous power model is formed by coupling air resistance power, slope resistance power and inertia power.

[0035] Wherein, the air resistance power ; Slope resistance power ; The inertia power ; but .

[0036] Where, Indicates the static mass of the train; represents the gravitational acceleration constant.

[0037] Operating parameters: speed , acceleration , load and slope angle ; Environmental parameters: air density , air resistance coefficient and windward area ; Energy efficiency parameters: traction system efficiency .

[0038] Further: in, As the traction motor load changes, the traction system data comes from the real-time feedback of the onboard control system; Indicates the train traction power (unit: kW); Represents an infinitesimal time increment, used as the time differential unit in the integration process; Indicates the tracking start time; Represents the integration time variable.

[0039] Where, Indicates time The instantaneous carbon emission factor (unit: kg CO2 / kWh).

[0040] The edge prediction module 23 predicts the carbon emission value based on the carbon emission prediction model And output Boolean exception flag by introducing an auxiliary branch for exception detection .

[0041] Specifically, the edge prediction module 23 involves the following steps: A carbon emission prediction model is constructed using a long short-term memory network (LSTM) and an attention mechanism prediction model to dynamically predict carbon emission levels in the future. The preprocessed standardized data is used as the input of the carbon emission prediction model, and the predicted carbon emission value is output. ; An anomaly detection auxiliary branch constructed using binary classification logic is introduced into the carbon emission prediction model to output a Boolean anomaly flag. and prediction confidence values ; Generate prediction confidence values ​​based on the criteria automatically learned during model training ; Deep modeling is performed on the operating status and carbon emission trajectory in historical time series data, and an auxiliary binary classification result is output, that is, a prediction mark of whether there is an abnormal carbon emission at the current moment. The auxiliary output result, together with the timestamp, geographic location, and carbon emission value, constitutes a data structure that can be put on the chain.

[0042] At the same time, to support the closed-loop design of edge-side decision-making and on-chain collaboration, the calculation logic of the carbon emission prediction model is decentralized to the edge computing node and deployed locally together with the power modeling and carbon emission calculation modules. This fully utilizes the computing power of edge resources, reduces response latency, and improves system real-time performance and elasticity. Finally, the carbon emission prediction model packages the regression results and classification results into a triplet .

[0043] Where, Indicates the carbon emission value of the edge node in the future period predicted based on the current orbit status; Indicates whether the carbon emission prediction model determines that abnormal carbon emission behavior will occur during this period; Indicates the confidence value of the predicted abnormal event, which is used to determine the credibility or alarm trigger threshold.

[0044] The edge node completes the above structure generation based on standardized multi-source data, and synchronizes the triples through the off-chain data synchronization mechanism. The summary data structure is uploaded to the blockchain system and used as an input parameter of the blockchain early warning module 32 for judgment on the chain.

[0045] In this embodiment: The preprocessed standardized data in the edge computing unit 2 is organized into a time series window of fixed length and fed into the carbon emission prediction model.

[0046] In the Long Short-Term Memory (LSTM) network, for the first Characteristics of a moment , where the feature A data vector representing each moment in the input sequence (a multidimensional feature vector concatenated from preprocessed and standardized data, including speed, energy consumption, slope, carbon emissions, etc.).

[0047] First, use the forget gate to determine which historical information needs to be retained at the current moment; Generate updates to new information through the input gate and candidate memory; Fuse the results of the forget gate and the input gate to update the internal memory state; Finally, the hidden state at that moment is generated through the output gate ; Repeat the above steps until all moments are traversed to obtain the hidden state sequence of the entire window .

[0048] For each moment of hidden state Calculating attention scores , used to characterize the importance of this moment to the final prediction; All attention scores Normalize and get a set of attention weights ; Use the attention weight to perform weighted summation on the hidden state sequence to obtain a comprehensive context vector ; The context vector Input a linear regression layer to get the carbon emission prediction value at future time .

[0049] In the same context vector A small neural network is added to the above for binary classification prediction: First, a nonlinear mapping is performed (such as a fully connected layer with ReLU activation) to extract discriminant features; Then the Sigmoid function outputs a probability value between 0 and 1. , used to indicate the confidence level of the existence of abnormal carbon emission trend in the current period; The probability Compare with the preset threshold (0.7-0.8) and get a Boolean exception flag : If the probability value If it is not lower than the threshold, it is marked as abnormal, otherwise it is marked as normal; Finally, the carbon emission prediction model packages the regression results and classification results into a triplet: ; Where, Represents the prediction stage of the carbon emission prediction model, which is used to predict the future time Anomaly predictions; Represents the confidence value of the abnormal probability in the carbon emission prediction model.

[0050] This triplet structure includes accurate predictions of future carbon emissions, abnormal judgments at future moments, and confidence values ​​of abnormal probabilities, which can provide effective data support for downstream on-chain evidence storage and early warning logic.

[0051] In this embodiment, the railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing also includes a blockchain network unit 3, which is used to compress the cumulative carbon emission data and generate a Merkle tree root hash on the chain, and store the carbon emission data on the blockchain.

[0052] Among them, the blockchain network unit 3 includes a blockchain storage module 31 and a blockchain early warning module 32; The blockchain storage module 31 is used to perform hash compression on the carbon emission data from the edge computing unit 2 according to the time window, construct a dynamic shard Merkle tree and write it into the blockchain to ensure that the data is tamper-proof and traceable.

[0053] Among them, the Merkle tree adopts a dynamic sharding structure, and the sharding logic is based on the train operation interval division.

[0054] The blockchain early warning module 32 determines the threshold value based on the carbon emission data stored on the chain and the threshold judgment algorithm and triggers the early warning event, providing timely response for operation and maintenance and scheduling.

[0055] In this embodiment, when a high-speed train departs from section A, passes through section B, and finally arrives at section C, the system needs to dynamically track the carbon emission data of the train in the three operating sections and implement sharded evidence storage through blockchain.

[0056] Each operating interval (such as interval A) corresponds to an independent data shard, which stores the raw carbon emission data of the interval.

[0057] When a train crosses a section boundary (such as entering the track under the jurisdiction of section B), a new shard is automatically created.

[0058] If the amount of data in a certain interval is too large (for example, more than 10,000 records), it needs to be further divided into sub-shards.

[0059] Specifically, the rules for dynamic sharding of the operating interval are as follows: The railway network consists of The tracks between adjacent stations constitute a basic operating range. : ; in, Indicates the starting station of the interval; Indicates the terminal station of the section; The minimum geographic coordinate boundary (longitude and latitude) of the interval, ; Indicates the maximum geographic coordinate boundary (longitude and latitude) of the interval, ; Indicates the current operating range Minimum longitude of geographic boundaries; Indicates the current operating range the maximum longitude of the geographic boundary; Indicates the current operating range Minimum latitude of geographic boundaries; Indicates the current operating range The maximum latitude of a geographic boundary.

[0060] Each interval Corresponding to a dynamic shard : ; Where, Represents the shard key, which is determined by the running interval Location and time window Sure; Represents the carbon emission dataset stored in the shard, satisfying , its location And timestamp ; Indicates the initial capacity threshold of the shard (e.g. 1000 records); Indicates the shard split threshold (e.g. 10,000 records); Indicates the shard merge threshold (such as 200 records).

[0061] Based on train location and time window Determine the shard key , ensure that the shards are aligned with the operating range and data timing: ; in, represents the longitude of the train when the event occurred, represents the latitude of the train when the event occurred; The granularity of spatial fragmentation used to divide the longitude direction is to divide the entire interval longitude range into several small grid steps; Indicates the granularity of spatial fragmentation in the latitudinal direction; Indicates the timestamp of the current moment; Indicates the time partition window length (for example, 300 seconds or 5 minutes), which is used for time slicing. Is the exclusive OR operator symbol.

[0062] When the train enters a new section When the shard is created, the real-time location (latitude and longitude coordinates) and timestamp are continuously checked to see if they meet the new shard creation logic. If so, a new shard is triggered. create.

[0063] New Shard The creation logic is: The longitude and latitude of the train's current location are not within the geographical coverage of any existing shard, that is, it does not belong to any current shard. of and When the interval is Indicates the Valid shard instances; The current time exceeds the time window of the current shard (e.g. 5 minutes), even if the trains are still in the same geographical area, a new shard needs to be created; Create a new shard , the initial capacity is (e.g. 1000 records), the status is set to Active; Original shard (Such as the shards in the previous interval) are converted to Frozen (read-only); When sharding The number of data records stored in exceeds the split threshold , the split operation is triggered.

[0064] The specific splitting operation is: Will The data is divided into two parts according to the timestamp: First half Store in sub-shard ; Second half Store in sub-shard ; The sub-shard inherits the geographic scope of the original shard and time window , only the time boundary is updated; Where, Represents a dynamic shard The starting timestamp, that is, the time when the shard starts recording carbon emission data; Represents a dynamic shard The timestamp at which the record ends, that is, the time of the last data entry before the shard is frozen or archived; Indicates the middle time point used for shard splitting operation, that is, the time boundary for splitting shard data.

[0065] When adjacent shards and When low load and physical proximity are met at the same time, the merge operation is triggered and adjacent shards and Merge into (representing a new shard), whose geographical range covers the original two intervals, and the capacity is reset to , the original shard and Mark as Archived.

[0066] The low load situation is as follows: the data volume of both shards is lower than the merge threshold (such as 200), that is and .

[0067] Physical proximity means that the operating intervals of the two shards are physically adjacent. The terminal yes Starting point .

[0068] In this embodiment, for each slice The life cycle states are as follows: Active: The only shard to which data can be written, corresponding to the current operating section of the train; Frozen (read-only): After the train leaves the section, the data becomes read-only, and the complete data is retained for query; Archived: Shards are marked as archived after being split or merged, and only historical tracing is supported.

[0069] Furthermore, the specific steps involved in constructing the Merkle tree are: For each shard Carbon emissions data block in (Carbon Emissions Data Block Contains timestamp, location, carbon emissions and auxiliary flags , wherein the actual carbon emissions are stored by the blockchain storage module 31 (Actual carbon emissions is calculated by the distributed computing module 22) and the auxiliary mark Save as data block , and generate a historical exception flag , where if the model flag detects an anomaly or the threshold judgment detects an anomaly, then is equal to 1; otherwise, Equal to 0, the historical exception flag generated The sequence is permanently recorded on the chain), and the leaf node hash is calculated : in, Indicates that the train is The longitude of the data point; Indicates that the train is The latitude of the data point; Indicates the The instantaneous carbon emissions of each data point.

[0070] Hash all leaf nodes in the shard Sort by time and generate parent node hashes in pairs. If the number of leaf nodes is odd, the last node replicates itself and participates in the calculation. Repeat the above steps until a unique shard root hash is generated. : ; .

[0071] Hash all shard roots As a leaf node, build a global Merkle tree: .

[0072] Where, Indicates the A carbon emission data block, including timestamp, location, and carbon emission amount; Represents a data block timestamp, synchronized with the GNSS clock; Represents the Secure Hash Algorithm 256-bit, which is used to generate unique hash values ​​of fixed length; Indicates the hash value of the left child node of the current node; Indicates the hash value of the right child node of the current node; Represents the hash value of the parent node generated by concatenating the hashes of the left and right child nodes; Indicates fragmentation Merkle tree root hash; Indicates that for each carbon emission data block in the process of building the Merkle tree The calculated hash value; Indicates fragmentation The last carbon emission data block in The leaf node hash value; Represents the root hash of the global Merkle tree, which is constructed from the root hashes of all shards; Indicates the total number of shards; Indicates the The Merkle root hash of each shard; Represents the Merkle tree construction function, which generates the root node by layer-by-layer hash calculation. The input is an ordered set of hashes of all leaf nodes in the shard, and the output is the root hash value.

[0073] Shard root hash for each shard Participates in the global Merkle tree construction as an independent unit, and supports the dynamic creation, merging, or splitting of shards (such as generating a new shard when a train enters a new section).

[0074] When a new shard is created or an old shard is modified, the global Merkle tree only needs to update the affected branches.

[0075] In this embodiment: The specific steps involved in data sharding are: Interval A: Generate shards , stores data from 14:00 to 14:30, a total of 800 records; Section B: The train enters a new section and creates , initial capacity 1000; Interval C: Due to the surge in data volume (1500 records), the shard split is triggered 、 .

[0076] The specific steps involved in hashing evidence: Shard root hash: Global root hash: Where, Indicates fragmentation Root hash of Indicates the The hash value of each (e.g., the 1st, 2nd, 800th) carbon emission data block; Represents the root hash calculated independently for each shard (such as A, B, C1, C2), which represents the summary of all data records in the shard.

[0077] Extract interval B data: Provide data blocks Merkle path: ; Where, Representation and data blocks A sibling node at the same level (i.e., another child node in the tree structure used to calculate its parent node); Represents the set of nodes that merge with the current branch to form the intermediate hash path; Represents the root hash of shard B, which contains the data block intermediate goals; Represents the global root hash, which will eventually be matched with the hash value recorded on the chain to confirm that the data has not been tampered with.

[0078] Calculate hash layer by layer to confirm Consistent with the on-chain records.

[0079] In this embodiment, in the blockchain early warning module 32, the specific steps involved in the threshold judgment algorithm judging the threshold and triggering the early warning event are: Based on the blockchain early warning module 32 monitoring chain newly added shard root hash ; Locate the carbon emission dataset within the shard through on-chain pointer index and Merkle path ; Decompress the carbon emission data block ; The above Indicates the flag used in the on-chain evidence storage stage, indexed at the actual arrival time of a data point (For example, when a new carbon emission record is written or detected in the shard When the record is abnormal, the judgment on whether it is abnormal is written to the chain), which represents the evidence storage stage, Abnormal flag; Carbon emission prediction model predicts the output of Boolean abnormal flag After being put on the chain or threshold verification, it becomes on the chain (Save as ).

[0080] Based on vehicle operating range Dynamically adjust the threshold based on the difficulty coefficient (such as slope, weather); Iterate over the carbon emissions dataset For each data, if the actual carbon emissions Exceeding the set threshold is recorded as an abnormal event ; At the same time, when new abnormal events are to be evaluated (for example, in the carbon emission dataset within the shard New data block encountered in ), looking back to the most recent On-chain at this moment sequence: Each Contains data from edge models result, and Implicit inheritance models the output of auxiliary branches.

[0081] Based on the threshold judgment algorithm and historical on-chain data, determine whether there is an anomaly (continuous anomaly or isolated anomaly): Where, Indicates the Abnormal carbon emission events detected at any time; Indicates the The actual carbon emission value detected at any moment; Indicates the current train running section Historical average carbon emissions; Indicates at time The on-chain continuous anomaly probability coefficient is a weighted indicator of the occurrence of continuous anomalies in the historical period, reflecting whether the current anomaly is in a period of continuous anomaly (a value close to 1 is more likely to be a continuous anomaly); Indicates at time The isolated anomaly penalty coefficient is used to suppress the impact of accidental or sporadic mutation events; Represents the weight coefficient of the historical deviation term, which is used to measure the impact intensity of the difference between the current emission value and the historical mean; represents the continuous abnormality discrimination weight coefficient; represents the isolated anomaly discrimination weight coefficient, which is used to control the influence of sudden outliers on the judgment results; and .

[0082] Further, Indicates the length of the time window; Indicates time Whether an abnormality occurs at the moment, (1 means there is an abnormality, 0 means there is no abnormality); Indicates the The timestamp of the data, i.e. the time when the abnormal carbon emission event occurred, is synchronized with the GNSS global clock; Indicates the explanation of the cause or basis for the judgment of the triggering of the abnormal event, which provides decision support for subsequent warnings or scheduling responses (including but not limited to the abnormality exceeding the set threshold, the Boolean abnormality flag output by the edge model, and the Boolean abnormality flag output by the edge model); Its value range is , among which, if If it is close to 1, it means that abnormalities have occurred frequently recently and the current abnormality is a persistent abnormality. If it is close to 0, it means there are almost no abnormalities in the recent period and the current abnormality is isolated. This helps to determine the background of the current abnormality and prevent accidental sudden points from being misjudged. Represents the penalty weight coefficient (0.5-0.8), which is used to control the weight of isolated anomalies; represents the longitude of the train when the event occurred; Indicates the latitude of the train when the event occurred.

[0083] like Exceeding the preset alarm threshold , the exception enters the summary data structure .

[0084] Among them, abnormal events are confirmed and encapsulated into a summary data structure as off-chain warning records. ; The summary data structure The fingerprint is uploaded to the chain (or stored in the off-chain database and uploaded to the chain), and sent to the dispatch center synchronously to trigger operation and maintenance responses (such as deceleration, temporary parking, load adjustment, etc.).

[0085] In this embodiment, the railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing also includes a dynamic tracking unit 4, which corrects the on-chain data and real-time edge computing data based on the LSTM prediction model, and uses the NSGA-II multi-objective genetic algorithm with elite retention strategy to perform global optimization and generate a Pareto solution set. , and the Pareto solution set Sent to edge computing unit 2 for closed-loop optimization.

[0086] Specifically, the dynamic tracking unit 4 includes a collaborative correction module 41 , a multi-objective optimization decision module 42 , a full life cycle carbon integration module 43 , and a closed-loop strategy issuing module 44 .

[0087] Among them, the collaborative correction module 41 is based on the summary data structure , triples and historical carbon emission series, and the LSTM prediction model is used to predict the efficiency of the traction system. and real-time emission factors Perform short-term prediction and correction to modify edge parameters.

[0088] In this embodiment, the input of the LSTM prediction model includes: Summary Data Structure , triples and historical carbon emission series ; Calibrate the real-time emission factor predicted by LSTM based on on-chain historical data : Where, Represents the weight coefficient, which is used to balance the influence of the LSTM model prediction value and the historical data on the chain. The value range is [0,1]; Indicates the calibration window length (default seconds), the larger the window, the more stable the calibration but the slower the response; Indicates that the LSTM model is at the current moment The predicted instantaneous emission factors are not corrected by on-chain data; Indicates a point in history The actual instantaneous carbon emissions; Indicates at a point in time Instantaneous energy consumption; represents the calibrated emission factor.

[0089] Traction system efficiency Compensation Mechanism: When the efficiency prediction value reported by the edge node deviates from the measured value by more than 5%, a sliding window variance analysis (window = 10 seconds) is started: Where, Represents the attenuation factor (default ); represents the traction system efficiency obtained by edge prediction at the current moment; It represents the efficiency deviation between the predicted value and the measured value; It represents the traction system efficiency after dynamic correction, which serves as the input value for subsequent carbon emission calculation and power modeling.

[0090] If a persistent positive deviation is detected, the predicted efficiency value is adjusted downwards exponentially; If a persistent negative deviation is detected, the predicted efficiency value is adjusted upward exponentially.

[0091] The multi-objective optimization decision module 42 takes edge computing delay, energy consumption, and link overhead as the objective function, and comprehensively considers the sampling interval , edge computing quota and chain frequency , using NSGA-II multi-objective genetic algorithm with elite retention strategy to perform global optimization and generate Pareto solution set , providing the system with the best operating solution under multiple performance trade-offs, among which, Indicates the The sampling interval of the solution; Indicates the The proportion of edge computing resources in the solution; Indicates the The frequency of uplinking of the solution; represents the number of Pareto solutions (indicates the number of optimal combinations that satisfy different performance preferences); Represents an index variable.

[0092] In this embodiment, the sampling interval Second; Edge computing quota CPU; Chain frequency Hz.

[0093] The NSGA-II multi-objective genetic algorithm with elite retention strategy is used to perform global optimization and generate Pareto solution sets. The specific steps involved are: Initialize the population (population size ), the individual coding form is The triplet of Generate Pareto frontier solution set through non-dominated sorting; The crowding distance operator is used to maintain the diversity of the solution set; Output optimal solution set , the optimal solution set Include Group parameter combinations that meet different performance preferences , according to the real-time network status and edge node resource margin, from the optimal solution set Select the operation strategy that suits the current scenario.

[0094] The full life cycle carbon integration module 43 calculates the comprehensive carbon emissions based on the carbon footprint expansion model If comprehensive carbon emissions are detected When the preset threshold is exceeded, the on-chain carbon transaction is triggered, where the carbon footprint expansion model is based on the cumulative carbon emissions , which is constructed by introducing the indirect carbon emission factors generated during the use of infrastructure.

[0095] In this embodiment, the carbon footprint expansion model is based on the cumulative carbon emissions By introducing indirect carbon emission factors generated during the use of infrastructure and performing weighted integration, a full life cycle carbon emission model is constructed: Where, Indicates time Comprehensive carbon emissions over its entire life cycle (in kg CO2); Indicates the total number of facility types, referring to the total number of infrastructure participating in carbon allocation (such as tracks, signal systems, power supply stations, etc.); Indicates the Unit life cycle carbon emissions of this type of facility (in kg CO2); Indicates the Carbon weight coefficient of type facility (by train to facility The proportion of occupied time); in, , where Indicates the train operation task to the facility The actual occupation time (unit: h); Display facilities Design life cycle (unit: h).

[0096] Cumulative carbon emissions It is used to characterize the direct carbon emissions generated by energy consumption during the operation of the train and is the basic parameter for constructing the carbon footprint model of the train life cycle.

[0097] A carbon footprint extension model is constructed to dynamically estimate the total amount of indirect carbon emissions based on the infrastructure life cycle and the train occupancy ratio, and trigger the on-chain carbon trading mechanism when the threshold is exceeded, to achieve the integrated integration of multi-source carbon factors and the carbon efficiency evaluation of the operation strategy, and solve the problem of isolated emission assessment during the operation stage and difficulty in connecting with the carbon market; specifically, the full life cycle carbon integration module 43 is used to further integrate the indirect carbon emission factors at the infrastructure level on the basis of completing the dynamic calculation of carbon emissions during the train operation process, to construct a full life cycle carbon emission model, and to enhance the system's perception and management capabilities of the carbon footprint of the entire railway transportation process.

[0098] The closed-loop strategy issuing module 44 is used to set the Pareto solution Sent to edge computing unit 2.

[0099] Example 2: The difference between Example 2 and Example 1 of the present invention is that this example introduces a railway transportation carbon footprint dynamic tracking method based on blockchain and edge computing used in a railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing: A method for dynamically tracking the carbon footprint of railway transportation based on blockchain and edge computing, used in any of the above-mentioned systems for dynamically tracking the carbon footprint of railway transportation based on blockchain and edge computing, comprising the following steps: S10.1. The edge computing unit 2 integrates the multi-source data collected by the data acquisition unit 1, calculates the real-time energy consumption of the train in combination with the instantaneous power model, introduces a dynamic emission factor, and converts it into cumulative carbon emissions.

[0100] S10.2. Blockchain network unit 3 compresses and encodes data from edge computing, constructs a hash structure of carbon emission data blocks through a dynamic sharding Merkle tree, and uploads it to the blockchain, achieving traceable evidence. When a train crosses an operating range or time window, it automatically triggers the creation, splitting, or merging of new shards. At the same time, by monitoring the newly added shard hash on the chain, abnormal events are identified based on the threshold judgment algorithm combined with historical data, and the abnormal summary data structure is uploaded to the chain and pushed to the dispatch center to trigger an operation and maintenance response.

[0101] S10.3. The on-chain data and edge prediction results are collaboratively corrected through the dynamic tracking unit 4. The NSGA-II multi-objective optimization algorithm is used to generate the Pareto solution set. After comprehensively considering the delay, energy consumption and on-chain overhead, the optimal strategy is fed back to the edge computing unit by the closed-loop strategy distribution module to achieve full-process optimization.

[0102] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing, characterized by: include: A data acquisition unit (1), the data acquisition unit (1) is used to collect train running status, environmental parameters and traction energy consumption data; An edge computing unit (2) is configured to fuse the multi-source data collected by the data collection unit (1) through an adaptive Kalman filter to obtain standardized data, and to convert the train operation status into carbon emissions based on an instantaneous power model, and to introduce a dynamic emission factor. Calculate cumulative carbon emissions ; A blockchain network unit (3), wherein the blockchain network unit (3) is used to compress the accumulated carbon emission data to generate a Merkle tree root hash and store the carbon emission data on the blockchain; Dynamic tracking unit (4) uses the LSTM prediction model to correct the on-chain data and real-time edge computing data, and uses the NSGA-II multi-objective genetic algorithm with elite retention strategy to perform global optimization and generate the Pareto solution set. , and the Pareto solution set The data is sent to the edge computing unit (2) for closed-loop optimization.

2. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 1 is characterized in that: The data acquisition unit (1) includes an on-board sensor module, a track environment monitoring module and an energy consumption metering module; Among them, the on-board sensor module is used to monitor the train's energy consumption, load, speed and location data; The track environment monitoring module is used to collect track slope, weather conditions, and signal system status parameters; The energy consumption metering module is used to collect energy consumption data of traction power supply systems and stations.

3. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 1 is characterized in that: The edge computing unit (2) includes a local pre-processing node module (21), a distributed computing module (22) and an edge prediction module (23); The local pre-processing node module (21) uses an adaptive Kalman filter to fuse the multi-source data collected by the data collection unit (1) to obtain pre-processed standardized data; The distributed computing module (22) converts the train operation status into carbon emissions using an instantaneous power model based on the pre-processed standardized data; Among them, the train power at time is calculated based on the instantaneous power model. The instantaneous total power and increase the efficiency of the traction system Introduced instantaneous total power Get cumulative energy consumption , and based on real-time emission factors The accumulated energy consumption Converted into cumulative carbon emissions; The instantaneous power model is formed by coupling air resistance power, slope resistance power and inertia power; The edge prediction module (23) predicts the carbon emission value based on the carbon emission prediction model And output Boolean exception flag by introducing an auxiliary branch for exception detection .

4. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 3 is characterized in that: The edge prediction module (23) involves the following specific steps: A carbon emission prediction model is constructed using a long short-term memory network and an attention mechanism prediction model to dynamically predict carbon emission levels in future periods. The preprocessed standardized data is used as the input of the carbon emission prediction model, and the predicted carbon emission value is output. ; An anomaly detection auxiliary branch constructed using binary classification logic is introduced into the carbon emission prediction model to output a Boolean anomaly flag. and prediction confidence values ; Finally, the carbon emission prediction model packages the regression results and classification results into a triplet .

5. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 1 is characterized in that: The blockchain network unit (3) includes a blockchain storage module (31) and a blockchain early warning module (32); The blockchain storage module (31) is used to perform hash compression on the carbon emission data from the edge computing unit (2) according to the time window, construct a dynamic sharding Merkle tree, and write the data into the blockchain; The Merkle tree adopts a dynamic sharding structure, and the sharding logic is based on the train operation interval. The blockchain early warning module (32) determines the threshold value and triggers an early warning event based on the carbon emission data stored on the chain and the threshold judgment algorithm.

6. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 5 is characterized in that: The dynamic sharding rules involved in the train running section are as follows: The railway network consists of The tracks between adjacent stations constitute a basic operating range. ; Each interval Corresponding to a dynamic shard ; Based on train position and time window Determine the shard key ; When the train enters a new section When the shard is created, the real-time location and timestamp are continuously checked to see if they meet the new shard creation logic. If they do, a new shard is triggered. create; When sharding The number of data records stored in exceeds the split threshold When , the split operation is triggered; When adjacent shards and When low load and physical proximity are met at the same time, the merge operation is triggered and adjacent shards and Merge into .

7. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 6 is characterized in that: The specific steps involved in constructing the Merkle tree are: For each shard Carbon emissions data block in , calculate the leaf node hash ; Hash all leaf nodes in the shard Sort by time and generate parent node hashes in pairs. Repeat the above steps until a unique shard root hash is generated. ; Hash all shard roots As a leaf node, build a global Merkle tree.

8. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 5 is characterized in that: In the blockchain early warning module (32), the specific steps involved in the threshold judgment algorithm judging the threshold and triggering the early warning event are: Based on the blockchain early warning module (32) monitoring the newly added shard root hash on the chain ; Locate the carbon emission dataset within the shard through on-chain pointer index and Merkle path ; Iterate over the carbon emissions dataset For each data, if the actual carbon emissions Exceeding the set threshold is recorded as an abnormal event ; At the same time, when new abnormal events are to be evaluated, the most recent On-chain at this moment sequence; Based on the threshold judgment algorithm and historical on-chain data, determine whether there is an abnormality: Where, Indicates the Abnormal carbon emission events detected at any time; Indicates the The actual carbon emission value detected at any moment; Indicates the current train running section Historical average carbon emissions; Indicates at time The probability coefficient of continuous anomalies on the chain; Indicates at time The isolated anomaly penalty coefficient; Represents the weight coefficient of historical deviation term; represents the continuous abnormality discrimination weight coefficient; represents the isolated anomaly discrimination weight coefficient; like Exceeding the preset alarm threshold , the exception enters the summary data structure ; Among them, abnormal events are confirmed and encapsulated into a summary data structure as off-chain warning records. ; The summary data structure The data is uploaded to the chain and sent synchronously to the dispatch center to trigger an operation and maintenance response.

9. The railway transportation carbon footprint dynamic tracking system based on blockchain and edge computing according to claim 1 is characterized in that: The dynamic tracking unit (4) includes a collaborative correction module (41), a multi-objective optimization decision module (42), a full life cycle carbon integration module (43), and a closed-loop strategy issuance module (44); Among them, the collaborative correction module (41) is based on the summary data structure , triples and historical carbon emission series, and the LSTM prediction model is used to predict the efficiency of the traction system. and real-time emission factors Perform short-term prediction and correction to modify edge parameters; The multi-objective optimization decision module (42) takes edge computing delay, energy consumption, and uplink overhead as the objective function, and comprehensively considers the sampling interval , edge computing quota and chain frequency , using NSGA-II multi-objective genetic algorithm with elite retention strategy to perform global optimization and generate Pareto solution set ; The full life cycle carbon integration module (43) calculates the comprehensive carbon emissions based on the carbon footprint expansion model If comprehensive carbon emissions are detected When the preset threshold is exceeded, the on-chain carbon transaction is triggered, where the carbon footprint expansion model is based on the cumulative carbon emissions , it is constructed by introducing the indirect carbon emission factors generated during the use of infrastructure; The closed-loop strategy issuing module (44) is used to set the Pareto solution Send it to the edge computing unit (2).

10. A method for dynamically tracking the carbon footprint of railway transportation based on blockchain and edge computing, used in a system for dynamically tracking the carbon footprint of railway transportation based on blockchain and edge computing as claimed in any one of claims 1 to 9, characterized in that: The steps include: S10.

1. The edge computing unit (2) fuses the multi-source data collected by the data collection unit (1), calculates the real-time energy consumption of the train in combination with the instantaneous power model, introduces a dynamic emission factor, and converts it into cumulative carbon emissions; S10.2, the blockchain network unit (3) compresses and encodes the data from the edge computing, constructs the hash structure of the carbon emission data block through the dynamic sharding Merkle tree and uploads it to the chain to achieve traceable evidence; when the train crosses the operating range or time window, it automatically triggers the creation, splitting or merging of new shards; At the same time, by monitoring the newly added shard hash on the chain, based on the threshold judgment algorithm combined with historical data to identify abnormal events, and then upload the abnormal summary data structure to the chain and push it to the dispatch center to trigger the operation and maintenance response; S10.

3. The on-chain data and edge prediction results are collaboratively corrected through the dynamic tracking unit (4). The NSGA-II multi-objective optimization algorithm is used to generate the Pareto solution set. After comprehensively considering the delay, energy consumption and on-chain overhead, the optimal strategy is fed back to the edge computing unit by the closed-loop strategy distribution module to achieve full-process optimization.

Images