A railway dispatching safety data aggregation and processing method based on AI model training

Abstract

This invention discloses a railway dispatching safety data aggregation and processing method based on AI model training, comprising the following steps: S1. Collecting railway dispatching-related data from multiple data sources, including train operation data, dispatching commands, track conditions, signal information, and meteorological data; cleaning and standardizing the collected data; S2. Integrating different types of data using multimodal fusion methods in deep learning; S3. Extracting key features from the fused dataset, using an autoencoder method for time series feature extraction, and using XGBoost for feature selection; S4. Using the processed dataset to train an AI model, selecting a random forest for training; and S5. Applying the trained AI model to real-time data. This invention provides solid technical support for improving the safety and intelligence level of railway dispatching systems.

Description

Technical Field

[0001] The present invention relates to the field of AI model technology, and in particular to a railway dispatching safety data aggregation and processing method based on AI model training. Background Art

[0002] Railway dispatching is a critical component of the railway transportation system, directly impacting the safety and efficiency of train operations. During the railway dispatching process, multiple factors, including train operation data, dispatching commands, track conditions, signal information, and weather data, must be comprehensively considered to ensure safe and on-time train operation. However, existing railway dispatching systems primarily rely on manual decision-making, which presents the following core difficulties:

[0003] 1. Numerous and fragmented data sources: Railway dispatching involves a variety of data sources, including train operating status, dispatching orders, track conditions, signal information, and meteorological data. This data is often scattered across different systems and lacks unified standards and formats, making effective comprehensive analysis difficult.

[0004] 2. Complex data processing: Due to the diversity and complexity of railway dispatching data, data cleaning, standardization, and aggregation are required to ensure data quality and consistency. This process typically requires extensive manual intervention and expertise, is time-consuming, labor-intensive, and prone to errors.

[0005] 3. Difficulty in feature extraction and selection: Analyzing railway dispatching data requires extracting and selecting key features that reflect dispatching safety from massive amounts of high-dimensional data. Existing methods, which rely heavily on traditional statistical methods and expert experience, struggle to fully exploit potential patterns and trends in the data, limiting prediction accuracy and model performance.

[0006] 4. High real-time requirements: Railway dispatching systems must be able to rapidly analyze and process real-time data, promptly predict potential safety hazards and dispatching risks, and provide corresponding decision support and safety warnings. Existing systems lack real-time and accuracy, failing to meet the requirements for safe and efficient railway transportation.

[0007] Therefore, how to provide a railway dispatching safety data aggregation and processing method based on AI model training to achieve comprehensive processing of multi-source data, efficient feature extraction and selection, real-time prediction and decision support has become a problem that technical personnel in this field urgently need to solve. Summary of the Invention

[0008] One purpose of the present invention is to propose a railway scheduling security data aggregation processing method based on AI model training.

[0009] According to an embodiment of the present invention, a railway dispatching safety data aggregation and processing method based on AI model training includes the following steps:

[0010] S1. Collect railway dispatch-related data from multiple data sources, including train operation data, dispatching orders, track conditions, signal information, and meteorological data, and clean and standardize the collected data;

[0011] S2. Use multimodal fusion methods in deep learning to integrate different types of data to form a unified data set;

[0012] S3, extract key features from the fused dataset, use the autoencoder method for time series feature extraction, and use XGBoost for feature selection;

[0013] S4. Use the processed data set to train the AI ​​model, select random forest for training, and optimize the model parameters through hyperparameter tuning technology;

[0014] S5. Real-time prediction and decision support: Apply the trained AI model to real-time data to make real-time predictions of safety hazards and scheduling risks, and provide decision support and safety warnings.

[0015] Optionally, S1 includes the following steps:

[0016] S11. Collecting railway dispatch-related data in real time from multiple data sources, including train operation data, dispatching commands, track conditions, signal information, and meteorological data;

[0017] S12. Express the collected train operation data as a time series x t , where x t represents the train position, speed and acceleration at time t;

[0018] S13, structurally process the dispatch command data and extract key information c in the dispatch command i , where c i Represents the content and timestamp of the i-th scheduling command;

[0019] S14, perform spatial mapping on the track condition data to generate a track condition matrix R = {r ij}, where r ij represents the status of the track segment i at the jth moment;

[0020] S15. Collect signal information data and convert it into a signal state sequence s t , where s t Represents the signal state at time t;

[0021] S16, obtain meteorological data, and perform spatial and temporal matching to form a meteorological condition matrix W = {w ij}, where w ij represents the meteorological conditions at the i-th location at the j-th time;

[0022] S17. Perform data cleaning on all collected data, including processing missing values, outliers, and noisy data. Use interpolation to process missing values ​​and use triple standard deviation method to detect and process outliers.

[0023] S18. Standardize the cleaned data and use the z-score standardization method to convert the data into a standard normal distribution with a mean of 0 and a standard deviation of 1.

[0024] Optionally, S2 includes the following steps:

[0025] S21, the pre-processed train operation data x t , scheduling command data c i , track condition data R = {r ij}、Signal information data t and meteorological data W = {w ij} Perform preliminary feature extraction to form a feature vector set f x ,f c ,f r ,f s ,f w , where x t represents the train position, speed and acceleration at time t, c i Represents the content and timestamp of the i-th scheduling command, r ij represents the status of the track segment i at the jth moment, s t represents the signal state at time t, w ij represents the meteorological conditions at the i-th location at the j-th time, f x 、f c 、f r 、f s 、f w : eigenvectors representing train operation data, dispatching command data, track condition data, signal information data, and meteorological data respectively;

[0026] S22. Use a deep neural network to encode each type of feature vector separately to form an intermediate representation h x , h c , h r , h s , h w , where DNN represents the deep neural network function, h x =DNN(f x) represents the intermediate representation of train operation data, h c =DNN(f c ) represents the intermediate representation of the scheduling command data, h r =DNN(f r ): represents the intermediate representation of track status data, h s =DNN(f s ) represents the intermediate representation of signal information data, h w =DNN(f w ) represents an intermediate representation of meteorological data;

[0027] S23. Use the self-attention mechanism to perform weighted fusion of different types of intermediate representations;

[0028] S24. Based on the calculated attention weights, perform weighted summation on the intermediate representations to form the fused feature representation H:

[0029] H=α x h x +α c h c +α r h r +α s h s +α w h w ;

[0030] Among them, H represents the fused feature representation;

[0031] S25, the fused feature representation H is processed through the fully connected layer to reduce the dimension and obtain a unified dataset representation {z k}, where: z k =FC(H);

[0032] Among them, z k Represents the feature representation after dimensionality reduction;

[0033] S26, for a unified dataset representation {z k} is normalized to form the final fusion data set {z′ k}.

[0034] Optionally, S3 specifically includes the following steps:

[0035] S31, the fused railway dispatching safety dataset z′ k Perform preliminary processing and construct the feature matrix F = {f kj}, where f k , represents the jth feature of the kth sample, the feature is the track condition, z′ krepresents the normalized kth sample, including train operation data, dispatching commands, track conditions, signal information and meteorological data, and F represents the feature matrix of all samples;

[0036] S32, use the autoencoder method to extract time series features from the feature matrix F, and build an autoencoder model, including the encoder and decoder parts. The encoder takes the input feature f kj Mapped to the hidden layer representation h k , the decoder represents the hidden layer h k Reconstruct back to the input features, the optimization goal is:

[0037]

[0038] Among them, θ represents the parameters of the autoencoder model, including the encoder parameters θ E and decoder parameters θ D , represents the reconstructed feature matrix, Where E(θ E , F) represents the output of the encoder, D(θ D ,E(θ E , F) represents the output of the decoder, || represents the norm of the reconstruction error, usually the L2 norm, λ represents the regularization parameter used to control the complexity of the model, L represents the number of layers of the autoencoder, W l represents the weight matrix of the lth layer, b l Represents the bias vector of the lth layer;

[0039] S33. Extracting time series features {h k}:

[0040] h k =E(θ E , f k );

[0041] Among them, h k Indicates the historical operation mode of the train, the change of track conditions, and the change of signal status. E(θ E , f k ) indicates that the encoder transforms the feature f of the kth sample k The process of mapping to the hidden layer;

[0042] S34. Use the XGBoost algorithm for feature selection and build a set of decision trees. The objective function is:

[0043]

[0044] Where Θ represents the parameters of the XGBoost model, N represents the number of samples, that is, the number of samples of railway scheduling safety data, L represents the loss function, which is used to measure the error between the predicted value and the true value, and y i represents the true label of the i-th sample, such as the safety status of train scheduling, represents the predicted value of the i-th sample, T represents the number of decision trees Ω(f t ) represents the regularization term of the t-th tree:

[0045]

[0046] Among them, γ represents the regularization parameter, which controls the complexity of the tree, d represents the number of features, and w j represents the weight of feature j;

[0047] S35. Calculate the importance score of each feature I j ;

[0048] S36, according to the importance score I j Sort the features, select the top n features with the highest scores, and construct the final feature set F′ to reflect the factors of railway scheduling safety, where n is the preset number of features;

[0049] S37. The final feature set F′ is used as input data for subsequent model training to further improve the accuracy and reliability of railway scheduling safety prediction.

[0050] Optionally, the S35 specifically includes:

[0051] S351, for each iteration t, using the current residual Build a new decision tree and update the model's predictions:

[0052]

[0053] in, represents the predicted value of the i-th sample after the t+1th round of iteration, represents the prediction of the railway scheduling safety data by the model after the t+1th round of iteration, η is the learning rate, which is used to control the update amplitude of the predicted value at each iteration, and f t (x i ) represents the predicted value of the t-th tree for sample i, and represents the contribution of the t-th decision tree to the railway scheduling safety data;

[0054] S352. In each round of iteration, the node splitting gain of the current tree is calculated. The gain formula is:

[0055]

[0056] Among them, Gs Represents the splitting gain of node s, which represents the information gain obtained by the current node when splitting, I L represents the sample set of the left child node, represents the railway scheduling safety data sample belonging to the left child node after splitting, I R represents the sample set of the right child node, represents the railway scheduling safety data samples belonging to the right child node after the split, I represents the sample set of the current node, represents all railway scheduling safety data samples before the split, g i Represents the first-order gradient of the i-th sample, which represents the derivative of the loss function to the predicted value and is used to measure the prediction error. i represents the second-order gradient of the i-th sample, represents the second-order derivative of the loss function with respect to the predicted value, and is used to measure the rate of change of the prediction error. λ represents the regularization parameter, which is used to control the complexity of the model and avoid overfitting.

[0057] S353. For each feature j, accumulate its split gain on all nodes and calculate the importance score I j :

[0058]

[0059] Among them, I j represents the importance score of the jth feature, which represents the importance of feature j in all decision trees, T represents the number of decision trees, which represents the total number of decision trees constructed, and S t represents the set of all nodes in the t-th tree, represents all nodes in the t-th decision tree, G js It represents the splitting gain of the jth feature at node s, which means the information gain obtained when feature j is split at node s;

[0060] S354. For each node s, the gain is calculated based on the accumulated values ​​of the first-order gradient and the second-order gradient. The calculation formula is:

[0061]

[0062] Among them, G js It represents the splitting gain of the jth feature at node s, and it represents the contribution of feature j when splitting at node s. s represents the sample set on node s, represents all railway scheduling safety data samples belonging to node s, g i Represents the first-order gradient of the i-th sample, which represents the derivative of the loss function to the predicted value and is used to measure the prediction error. i It represents the second-order gradient of the i-th sample and the second-order derivative of the loss function with respect to the predicted value, which is used to measure the rate of change of the prediction error.

[0063] Optionally, the S4 specifically includes:

[0064] S41, using the processed feature set F' to construct a training data set {(x i ,y i )}, where F′ represents the final feature set, which contains the features with the highest scores. The features represent the key information of railway scheduling safety data, y i represents the label of the i-th sample, indicating the safety status of railway scheduling, such as safety, potential risk or failure, x i The feature vector representing the i-th sample contains train operation data, dispatching commands, track conditions, signal information, and meteorological data;

[0065] S42, select the random forest algorithm for model training, and construct multiple decision trees {T m}, where each tree is trained using a different subset of features and samples:

[0066] {T m}={T1,T2,...,T M};

[0067] Among them, {T m} represents the set of decision trees in the random forest, T m : the mth decision tree, M: the number of decision trees, indicating the total number of trees contained in the random forest;

[0068] S43. Optimize model parameters through hyperparameter tuning technology to determine the best hyperparameter combination {θ k}, including the number of trees M, the maximum depth D, and the minimum number of sample splits N min :

[0069] (θ k}={θ1,θ2,...,θ K};

[0070] Among them, (θ k} represents the set of hyperparameters, θ k : The kth hyperparameter, K represents the number of hyperparameters, M represents the number of decision trees, which represents the total number of trees included in the random forest, D represents the maximum depth of the decision tree, which represents the maximum number of layers of the tree, and N min Indicates the minimum number of sample splits, which indicates the minimum number of samples required for a node to split;

[0071] S44. Use the grid search method to search for the best parameter combination in the preset parameter space. The optimization goal is:

[0072]

[0073] Among them, θ represents the hyperparameter to be optimized, N: the number of samples, which represents the total number of samples in the training data set, Loss function, which measures the predicted value and the true value y i the differences between;

[0074] S45. Evaluate model performance through cross-validation;

[0075] S46. Using the best hyperparameter combination Retrain the random forest model to get the final model

[0076]

[0077] in, The final random forest model represents the predicted value of the input feature vector x, T m (x): The predicted value of sample x by the mth decision tree. M represents the number of decision trees and the total number of trees contained in the random forest.

[0078] Optionally, the S45 specifically includes:

[0079] S451, the training data set {(x i ,y i )} is divided into K non-overlapping subsets, denoted as {D1, D2, ..., D K}, where the number of samples contained in each subset is

[0080] Among them, {(x i ,y i )} represents the training dataset, which contains the features and labels of railway scheduling safety data. K represents the number of cross-validation folds, and the training dataset is divided into K subsets. k represents the kth subset, each subset contains a part of railway scheduling safety data samples, N represents the total number of samples in the training data set, and represents the number of samples of railway scheduling safety data;

[0081] S452. Perform the following operations on each hyperparameter combination θ:

[0082] For the kth subset D k , which is used as the validation set, and the remaining K-1 subsets ({D1, D2, ..., D k-1 , D k+1 ,...,D K}Merge into training set:

[0083]

[0084] represents the training set in the kth fold, which is formed by merging the K-1 subsets and contains the features and labels of railway scheduling safety data;

[0085] represents the validation set in the k-th fold, which consists of the k-th subset and contains the features and labels of railway scheduling safety data;

[0086] In the training set Train the random forest model and get the model

[0087]

[0088] in, represents the predicted value of the input feature vector x obtained by the random forest model trained on the training set in the kth fold, and represents the prediction of railway scheduling safety data. represents the mth decision tree in the kth fold, the predicted value of the input feature vector x, represents the prediction of railway scheduling safety data, M represents the number of decision trees, and represents the total number of trees contained in the random forest;

[0089] Using a validation set Evaluate model performance and calculate loss function The average value of:

[0090]

[0091] Among them, L k represents the average loss in the kth fold, represents the model performance on the validation set, represents the number of samples in the validation set, Represents the model in the kth fold for the validation set sample x i The predicted value of Represents the loss function, which measures the predicted value and the true value y i The difference between them is used to evaluate the prediction accuracy of railway dispatching safety data;

[0092] S453. Calculate the average loss of all compromises as the performance indicator of the hyperparameter combination θ:

[0093]

[0094] Among them, L cv (θ) represents the cross-validation average loss of the hyperparameter combination θ, represents the model performance in all folds, K represents the number of cross-validation folds, L k represents the average loss in the kth fold;

[0095] S454. Select the hyperparameter combination with the minimum cross-validation average loss as the best hyperparameter combination:

[0096]

[0097] in, represents the best hyperparameter combination after optimization, represents the kth optimized hyperparameter, L cv (θ) represents the cross-validation average loss for the hyperparameter combination θ.

[0098] Optionally, the S5 specifically includes:

[0099] S51. Collect and pre-process real-time railway dispatching data, including train operation data, dispatching commands, track conditions, signal information and meteorological data, and convert these data into feature vectors x real-time ,

[0100] Among them, x real-time A feature vector representing real-time data, including the latest train status, dispatching commands, track conditions, signal information, and weather data;

[0101] S52, the real-time data feature vector x real-time Input to the trained AI model Generate real-time prediction results

[0102]

[0103] in, Indicates the prediction result of the AI ​​model on real-time data, indicating the safety status of the current railway scheduling. Represents a trained AI model used to predict the input feature vector x;

[0104] S53. Based on the prediction results Analyze safety hazards and scheduling risks, and determine whether there are potential risks based on preset thresholds:

[0105]

[0106] Among them, Risk represents the real-time risk level, which is classified according to the prediction results, τ high Indicates a high risk threshold, exceeding this value indicates high risk, τ low Indicates a low risk threshold, below which the risk is low;

[0107] S54. Provide corresponding decision support based on the risk level, including scheduling adjustment suggestions and safety measures suggestions. The specific steps are: when the risk level is high, take emergency scheduling adjustment measures immediately; when the risk level is medium, issue an early warning and recommend that the dispatcher take preventive measures; when the risk level is low, continue monitoring without taking additional measures.

[0108] The beneficial effects of the present invention are:

[0109] The railway dispatching safety data aggregation and processing method proposed in this paper, based on AI model training, comprehensively utilizes advanced algorithms such as computer deep learning and random forests, and describes in detail the specific steps for implementing railway dispatching safety data aggregation and processing. It has the following significant advantages:

[0110] 1. Strong data comprehensive processing capabilities: This method can effectively integrate multiple data sources such as train operation data, dispatching commands, track conditions, signal information and meteorological data through the aggregation and processing of multi-source data, and perform data cleaning, standardization and aggregation processing to ensure the comprehensiveness and consistency of data, thereby improving the accuracy and reliability of data analysis.

[0111] 2. Efficient feature extraction and selection: During the feature engineering process, this method uses an autoencoder for time series feature extraction and the XGBoost algorithm for feature selection. This method extracts key features from large amounts of high-dimensional data, improving the model's accuracy and generalization capabilities. Accurate feature selection enables the model to better capture important patterns and trends in the data.

[0112] 3. Advanced Model Training and Optimization: This method uses a random forest algorithm for model training and combines hyperparameter tuning techniques to optimize model parameters, ensuring high performance and stability. The application of cross-validation technology further improves the reliability of the model, providing a solid foundation for real-time prediction of railway dispatch safety.

[0113] 4. Real-time Prediction and Decision Support: This method applies a trained AI model to real-time data, enabling real-time prediction of safety hazards and scheduling risks, and providing corresponding decision support and safety warnings. By dynamically updating model input data, the real-time and accuracy of the railway dispatching system is ensured, helping dispatchers take timely measures to prevent potential accidents and improving the safety and efficiency of railway dispatching.

[0114] 5. High level of intelligence and automation: This invention uses advanced AI technology and data processing methods to realize the intelligent and automated processing of railway dispatching safety data, significantly reducing the dependence on manual experience, reducing the safety hazards and errors caused by human operation, and improving the overall intelligence level and operating efficiency of the railway dispatching system.

[0115] In summary, the present invention provides solid technical support for improving the safety and intelligence level of the railway dispatching system. It has the advantages of strong comprehensive data processing, accurate feature extraction, efficient model performance, reliable real-time prediction and high intelligence level, and can significantly improve the safety and efficiency of railway dispatching. BRIEF DESCRIPTION OF THE DRAWINGS

[0116] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0117] Figure 1 This is a flowchart of the overall method for railway dispatching safety data aggregation and processing based on AI model training proposed in this paper. The flowchart details the entire process from data collection and preprocessing, multimodal data fusion, high-dimensional feature extraction and selection, model training and optimization, to real-time prediction and decision support.

[0118] Figure 2 This is a detailed diagram of the data collection and preprocessing module, showing the steps of collecting, cleaning, and standardizing multi-source data such as train operation data, dispatching commands, track conditions, signal information, and meteorological data. DETAILED DESCRIPTION

[0119] The present invention will now be described in further detail with reference to the accompanying drawings, which are simplified schematic diagrams that illustrate the basic structure of the present invention in a schematic manner.

[0120] refer to Figure 1 , a railway dispatching safety data aggregation and processing method based on AI model training, comprising the following steps: S1, collecting railway dispatching-related data from multiple data sources, including train operation data, dispatching commands, track conditions, signal information and meteorological data, and cleaning and standardizing the collected data;

[0121] In this embodiment, S1 includes the following steps:

[0122] S11. Collecting railway dispatch-related data in real time from multiple data sources, including train operation data, dispatching commands, track conditions, signal information, and meteorological data;

[0123] S12. Express the collected train operation data as a time series x t , where x t represents the train position, speed and acceleration at time t;

[0124] S13, structurally process the dispatch command data and extract key information c in the dispatch command i , where ci Represents the content and timestamp of the i-th scheduling command;

[0125] S14, perform spatial mapping on the track condition data to generate a track condition matrix R = {r ij}, where r ij represents the status of the track segment i at the jth moment;

[0126] S15. Collect signal information data and convert it into a signal state sequence s t , where s t Represents the signal state at time t;

[0127] S16, obtain meteorological data, and perform spatial and temporal matching to form a meteorological condition matrix W = {w ij}, where w ij represents the meteorological conditions at the i-th location at the j-th time;

[0128] S17. Perform data cleaning on all collected data, including processing missing values, outliers, and noisy data. Use interpolation to process missing values ​​and use triple standard deviation method to detect and process outliers.

[0129] S18. Standardize the cleaned data and use the z-score standardization method to convert the data into a standard normal distribution with a mean of 0 and a standard deviation of 1.

[0130] S2. Integrate different types of data using a multimodal fusion method in deep learning to form a unified data set; optionally, S2 includes the following steps:

[0131] S21, the pre-processed train operation data x t , scheduling command data c i , track condition data R = {r ij}、Signal information data t and meteorological data W = {w ij} Perform preliminary feature extraction to form a feature vector set f x , f c , f r , f s , f w , where x t represents the train position, speed and acceleration at time t, c i Represents the content and timestamp of the i-th scheduling command, r ij represents the status of the track segment i at the jth moment, s t Indicates the signal state at time t, wi j represents the meteorological conditions at the i-th location at the j-th time, f x 、fc 、f r 、f s 、f w : eigenvectors representing train operation data, dispatching command data, track condition data, signal information data, and meteorological data respectively;

[0132] S22. Use a deep neural network to encode each type of feature vector separately to form an intermediate representation h x , h c , h r , h s , h w , where DNN represents the deep neural network function, h x =DNN(f x ) represents the intermediate representation of train operation data, h c =DNN(f c ) represents the intermediate representation of the scheduling command data, h r =DNN(f r ): represents the intermediate representation of track status data, h s =DNN(f s ) represents the intermediate representation of signal information data, h w =DNN(f w ) represents an intermediate representation of meteorological data;

[0133] S23. Use the self-attention mechanism to perform weighted fusion of different types of intermediate representations;

[0134] S24. Based on the calculated attention weights, perform weighted summation on the intermediate representations to form the fused feature representation H:

[0135] H=α x h x +α c h c +α r h r +α s h s +α w h w ;

[0136] Among them, H represents the fused feature representation;

[0137] S25, the fused feature representation H is processed through the fully connected layer to reduce the dimension and obtain a unified data set representation {z k}, where: z k =FC(H);

[0138] Among them, z k Represents the feature representation after dimensionality reduction;

[0139] S26, for a unified dataset representation {z k} is normalized to form the final fusion data set {z′ k}.

[0140] S3, extract key features from the fused dataset, use the autoencoder method for time series feature extraction, and use XGBoost for feature selection;

[0141] In this embodiment, S3 specifically includes the following steps:

[0142] S31, the fused railway dispatching safety dataset z′ k Perform preliminary processing and construct the feature matrix F = {f kj}, where f kj represents the jth feature of the kth sample, the feature is the track condition, z′ k represents the normalized kth sample, including train operation data, dispatching commands, track conditions, signal information and meteorological data, and F represents the feature matrix of all samples;

[0143] S32, use the autoencoder method to extract time series features from the feature matrix F, and build an autoencoder model, including the encoder and decoder parts. The encoder takes the input feature f kj Mapped to the hidden layer representation h k , the decoder represents the hidden layer h k Reconstruct back to the input features, the optimization goal is:

[0144]

[0145] Among them, θ represents the parameters of the autoencoder model, including the encoder parameters θ E and decoder parameters θ D , represents the reconstructed feature matrix, Where E(θ E , F) represents the output of the encoder, D(θ D ,E(θ E , F) represents the output of the decoder, || represents the norm of the reconstruction error, usually the L2 norm, λ represents the regularization parameter used to control the complexity of the model, L represents the number of layers of the autoencoder, W l represents the weight matrix of the lth layer, b l Represents the bias vector of the lth layer;

[0146] S33. Extracting time series features {h k}:

[0147] h k=E(θ E , f k );

[0148] Among them, h k Indicates the historical operation mode of the train, the change of track conditions, and the change of signal status. E(θ E , f k ) indicates that the encoder transforms the feature f of the kth sample k The process of mapping to the hidden layer;

[0149] S34. Use the XGBoost algorithm for feature selection and build a set of decision trees. The objective function is:

[0150]

[0151] Where Θ represents the parameters of the XGBoost model, N represents the number of samples, that is, the number of samples of railway scheduling safety data, L represents the loss function, which is used to measure the error between the predicted value and the true value, and y i represents the true label of the i-th sample, such as the safety status of train scheduling, represents the predicted value of the i-th sample, T represents the number of decision trees Ω(f t ) represents the regularization term of the t-th tree:

[0152]

[0153] Among them, γ represents the regularization parameter, which controls the complexity of the tree, d represents the number of features, and w j represents the weight of feature j;

[0154] S35. Calculate the importance score of each feature I j ;

[0155] S36, according to the importance score I j Sort the features, select the top n features with the highest scores, and construct the final feature set F′ to reflect the factors of railway scheduling safety, where n is the preset number of features;

[0156] S37. The final feature set F′ is used as input data for subsequent model training to further improve the accuracy and reliability of railway scheduling safety prediction.

[0157] S35. Calculate the importance score of each feature I j ;

[0158] In this embodiment, S35 specifically includes:

[0159] S351, for each iteration t, using the current residual Build a new decision tree and update the model's predictions:

[0160]

[0161] in, represents the predicted value of the i-th sample after the t+1th round of iteration, represents the prediction of the railway scheduling safety data by the model after the t+1th round of iteration, η is the learning rate, which is used to control the update amplitude of the predicted value at each iteration, and f t (x i ) represents the predicted value of the t-th tree for sample i, and represents the contribution of the t-th decision tree to the railway scheduling safety data;

[0162] S352. In each round of iteration, the node splitting gain of the current tree is calculated. The gain formula is:

[0163]

[0164] Among them, G s Represents the splitting gain of node s, which represents the information gain obtained by the current node when splitting, I L represents the sample set of the left child node, represents the railway scheduling safety data sample belonging to the left child node after splitting, I R represents the sample set of the right child node, represents the railway scheduling safety data samples belonging to the right child node after the split, I represents the sample set of the current node, represents all railway scheduling safety data samples before the split, g i Represents the first-order gradient of the i-th sample, which represents the derivative of the loss function to the predicted value and is used to measure the prediction error. i represents the second-order gradient of the i-th sample, represents the second-order derivative of the loss function with respect to the predicted value, and is used to measure the rate of change of the prediction error. λ represents the regularization parameter, which is used to control the complexity of the model and avoid overfitting.

[0165] S353. For each feature j, accumulate its split gain on all nodes and calculate the importance score I j :

[0166]

[0167] Among them, I j represents the importance score of the jth feature, which represents the importance of feature j in all decision trees, T represents the number of decision trees, which represents the total number of decision trees constructed, and S t represents the set of all nodes in the t-th tree, represents all nodes in the t-th decision tree, G js It represents the splitting gain of the jth feature at node s, which means the information gain obtained when feature j is split at node s;

[0168] S354. For each node s, the gain is calculated based on the accumulated values ​​of the first-order gradient and the second-order gradient. The calculation formula is:

[0169]

[0170] Among them, G js It represents the splitting gain of the jth feature at node s, and it represents the contribution of feature j when splitting at node s. s represents the sample set on node s, represents all railway scheduling safety data samples belonging to node s, g i Represents the first-order gradient of the i-th sample, which represents the derivative of the loss function to the predicted value and is used to measure the prediction error. i It represents the second-order gradient of the i-th sample and the second-order derivative of the loss function with respect to the predicted value, which is used to measure the rate of change of the prediction error.

[0171] S4. Use the processed data set to train the AI ​​model, select random forest for training, and optimize the model parameters through hyperparameter tuning technology;

[0172] In this embodiment, S4 specifically includes:

[0173] S41, using the processed feature set F ′ Construct training dataset {(x i ,y i )}, where F ′ represents the final feature set, which contains the features with the highest scores. The features represent the key information of railway scheduling safety data. i represents the label of the i-th sample, indicating the safety status of railway scheduling, such as safety, potential risk or failure, x i The feature vector representing the i-th sample contains train operation data, dispatching commands, track conditions, signal information, and meteorological data;

[0174] S42, select the random forest algorithm for model training, and construct multiple decision trees {T m}, where each tree is trained using a different subset of features and samples:

[0175] {T m}={T1,T2,...,T M};

[0176] Among them, {T m} represents the set of decision trees in the random forest, T m : the mth decision tree, M: the number of decision trees, indicating the total number of trees contained in the random forest;

[0177] S43. Optimize model parameters through hyperparameter tuning technology to determine the best hyperparameter combination {θ k}, including the number of trees M, the maximum depth D, and the minimum number of sample splits N min :

[0178] {θ k}={θ1,θ2,...,θ K};

[0179] Among them, {θ k} represents the set of hyperparameters, θ k : The kth hyperparameter, K represents the number of hyperparameters, M represents the number of decision trees, which represents the total number of trees included in the random forest, D represents the maximum depth of the decision tree, which represents the maximum number of layers of the tree, and N min Indicates the minimum number of sample splits, which indicates the minimum number of samples required for a node to split;

[0180] S44. Use the grid search method to search for the best parameter combination in the preset parameter space. The optimization goal is:

[0181]

[0182] Among them, θ represents the hyperparameter to be optimized, N: the number of samples, which represents the total number of samples in the training data set, Loss function, which measures the predicted value and the true value y i the differences between;

[0183] S45. Evaluate model performance through cross-validation;

[0184] S46. Using the best hyperparameter combination Retrain the random forest model to get the final model

[0185]

[0186] in, The final random forest model represents the predicted value of the input feature vector x, T m (x): The predicted value of sample x by the mth decision tree. M represents the number of decision trees and the total number of trees contained in the random forest.

[0187] S45. Evaluate model performance through cross-validation;

[0188] In this embodiment, S45 specifically includes:

[0189] S451, the training data set {(x i ,y i)} is divided into K non-overlapping subsets, denoted as {D1, D2, ..., D K}, where the number of samples contained in each subset is

[0190] Among them, {(x i ,y i )) represents the training dataset, which contains the features and labels of railway scheduling safety data. K represents the number of cross-validation folds, which divides the training dataset into K subsets. D k represents the kth subset, each subset contains a part of railway scheduling safety data samples, N represents the total number of samples in the training data set, and represents the number of samples of railway scheduling safety data;

[0191] S452. Perform the following operations on each hyperparameter combination θ:

[0192] For the kth subset D k , which is used as the validation set, and the remaining K-1 subsets ({D1, D2, ..., D k-1 , D k+1 ,...,D K}Merge into training set:

[0193]

[0194] represents the training set in the k-th fold, which is formed by merging the K-1 subsets and contains the features and labels of railway scheduling safety data;

[0195] represents the validation set in the k-th fold, which consists of the k-th subset and contains the features and labels of railway scheduling safety data;

[0196] In the training set Train the random forest model and get the model

[0197]

[0198] in, represents the predicted value of the input feature vector x obtained by the random forest model trained on the training set in the kth fold, and represents the prediction of railway scheduling safety data. represents the mth decision tree in the kth fold, the predicted value of the input feature vector x, represents the prediction of railway scheduling safety data, M represents the number of decision trees, and represents the total number of trees contained in the random forest;

[0199] Using a validation set Evaluate model performance and calculate loss function The average value of:

[0200]

[0201] Among them, L k represents the average loss in the kth fold, represents the model performance on the validation set, represents the number of samples in the validation set, Represents the model in the kth fold for the validation set sample x i The predicted value of Represents the loss function, which measures the predicted value and the true value y i The difference between them is used to evaluate the prediction accuracy of railway dispatching safety data;

[0202] S453. Calculate the average loss of all compromises as the performance indicator of the hyperparameter combination θ:

[0203]

[0204] Among them, L cv (θ) represents the cross-validation average loss of the hyperparameter combination θ, represents the model performance in all folds, K represents the number of cross-validation folds, L k represents the average loss in the kth fold;

[0205] S454. Select the hyperparameter combination with the minimum cross-validation average loss as the best hyperparameter combination:

[0206]

[0207] in, represents the best hyperparameter combination after optimization, represents the kth optimized hyperparameter, L cv (θ) represents the cross-validation average loss for the hyperparameter combination θ.

[0208] S5. Real-time prediction and decision support: Apply the trained AI model to real-time data to make real-time predictions of safety hazards and scheduling risks, and provide decision support and safety warnings.

[0209] In this embodiment, S5 specifically includes:

[0210] S51. Collect and pre-process real-time railway dispatching data, including train operation data, dispatching commands, track conditions, signal information and meteorological data, and convert these data into feature vectors x real-time ,

[0211] Among them, x real-time A feature vector representing real-time data, including the latest train status, dispatching commands, track conditions, signal information, and weather data;

[0212] S52, the real-time data feature vector x real-time Input to the trained AI model Generate real-time prediction results

[0213]

[0214] in, Indicates the prediction result of the AI ​​model on real-time data, indicating the safety status of the current railway scheduling. Represents a trained AI model used to predict the input feature vector x;

[0215] S53. Based on the prediction results Analyze safety hazards and scheduling risks, and determine whether there are potential risks based on preset thresholds:

[0216]

[0217] Among them, Risk represents the real-time risk level, which is classified according to the prediction results. high Indicates a high risk threshold, exceeding this value indicates high risk, τ low Indicates a low risk threshold, below which the risk is low;

[0218] S54. Provide corresponding decision support based on the risk level, including scheduling adjustment suggestions and safety measures suggestions. The specific steps are: when the risk level is high, take emergency scheduling adjustment measures immediately; when the risk level is medium, issue an early warning and recommend that the dispatcher take preventive measures; when the risk level is low, continue monitoring without taking additional measures.

[0219] Example 1:

[0220] In the actual railway dispatching system, it is crucial to ensure the safety of train operation and the effectiveness of dispatching. Let’s take the Beijing-Shanghai high-speed railway trunk line in June 2023 as an example. This trunk line needs to dispatch 300 trains every day. Dispatchers need to monitor a variety of data such as train operation status, track conditions, signal information, and meteorological data in real time to ensure the safe operation of trains. However, the existing system has significant deficiencies in data processing, feature extraction, model prediction, and decision support. By applying the railway dispatching safety data aggregation processing method based on AI model training of the present invention, we can significantly improve the safety and efficiency of railway dispatching.

[0221] In this embodiment, it is first necessary to collect and pre-process data from multiple data sources, including train operation data, dispatching commands, track conditions, signal information, and meteorological data. These data are acquired in real time by various sensors and monitoring systems and collected by the data acquisition module. For the data on June 10, 2023, we use data cleaning and standardization processing technology to remove noise and outliers, and convert the data into a unified standard format to ensure data quality and consistency. For example, in the data collection on that day, the train operation data involves train speed and position data recorded once a minute, totaling 1,440 records; the dispatching command data involves all dispatching instructions for the day, totaling 250 records; the track condition data includes track inspection data once a day and every hour, totaling 24 records; the signal information and meteorological data are recorded once a minute, totaling 1,440 records.

[0222] After data cleaning and standardization, we leverage multimodal fusion methods from deep learning to effectively integrate different types of data into a unified dataset. Specifically, we employ an autoencoder to extract features from the data. The encoder maps high-dimensional input data to a low-dimensional hidden layer representation. The decoder then reconstructs the input data and optimizes model parameters by minimizing the reconstruction error. In this process, the autoencoder automatically learns and extracts key features that reflect dispatch safety, thereby improving the accuracy of data analysis.

[0223] Next, we used the XGBoost algorithm for feature selection. The XGBoost algorithm constructs a gradient boosting decision tree, evaluates the importance of each feature, and selects the most representative features. For example, by analyzing data from June 10, 2023, the XGBoost algorithm identified key features such as the rate of change of train speed, the rate of change of track conditions, changes in signal status, and changes in weather conditions. These features have high split gains across multiple decision tree splits and can significantly impact the model's predictive performance.

[0224] After feature extraction and selection, we used the random forest algorithm to train the model. Using hyperparameter tuning techniques, we determined the optimal hyperparameter combination, including 100 decision trees, a maximum depth of 10, and a minimum number of sample splits of 5. In practice, we verified the model's performance through cross-validation to ensure high predictive accuracy and stability. Specific data demonstrated that the prediction accuracy of the training and validation sets reached 95% and 93%, respectively, demonstrating strong generalization capabilities.

[0225] By applying trained AI models to real-time data, we can predict dispatching safety hazards and risks in real time. For example, during a dispatch on June 11, 2023, the model analyzed the current train operating status, track conditions, and weather conditions in real time, predicting potential track risks for a particular train and issuing a warning signal when the risk level reached a high-risk threshold. Based on the warning information provided by the system, the dispatcher took timely scheduling adjustments, avoiding a potential accident.

[0226] Table 1 Data collection

[0227]

[0228]

[0229] Table 2 Feature extraction and selection results

[0230]

[0231] Table 3 Model training and optimization parameters

[0232] Parameter name Parameter Symbol Parameter value Number of decision trees M 100 Maximum depth D 10 Minimum number of sample splits N_min 5 Training set accuracy 95％ Validation set accuracy 93％

[0233] Table 4: Real-time prediction and decision support examples

[0234]

[0235]

[0236] Table 5: System performance comparison before and after application

[0237] index Before application After application Improve the situation accident rate 0.8％ 0.4％ 50% reduction Scheduling efficiency 70％ 91％ 30% increase Workload high middle Significantly reduced Data processing accuracy 80％ 95％ 15% increase

[0238] In conjunction with the above table, in order to verify the beneficial effects of the present invention, we compared the dispatching safety and efficiency before and after the adoption of the present invention. Before the adoption of the present invention, dispatchers needed to rely on experience to analyze data and make decisions, the accuracy and real-time performance of data processing were low, and safety hazards were not identified in a timely manner. After the adoption of the present invention, the system can automatically process and analyze large amounts of complex data, provide accurate predictions and decision support in real time, and greatly improve the safety and efficiency of dispatching. Specific data show that after the adoption of the present invention, the accident rate of the dispatching system dropped from 0.8% to 0.4%, the dispatching efficiency increased by 30%, the workload of the dispatchers was significantly reduced, and the safety and reliability were greatly improved.

[0239] In summary, this invention achieves intelligent and automated processing of railway dispatching safety data by effectively integrating multi-source data, employing advanced feature extraction and selection algorithms, and combining training and optimization with a random forest model. The application of this invention significantly improves the safety and efficiency of the railway dispatching system, providing a strong guarantee for the safe operation of railway transportation.

[0240] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A railway dispatching safety data aggregation and processing method based on AI model training, characterized in that: The steps include: S1. Collect railway dispatch-related data from multiple data sources, including train operation data, dispatching orders, track conditions, signal information, and meteorological data, and clean and standardize the collected data; S2. Use the multimodal fusion method in deep learning to integrate different types of data to form a unified data set, specifically including: S21. t , scheduling command data c i , track condition data R = {r ij }、Signal information data t and meteorological data W = {w ij } Perform preliminary feature extraction to form a feature vector set f x ,f c ,f r ,f s ,f w , where x t represents the train position, speed and acceleration at time t, c i Represents the content and timestamp of the i-th scheduling command, r ij represents the status of the track segment i at the jth moment, s t represents the signal state at time t, w ij represents the meteorological conditions at the i-th location at the j-th time, f x 、f c 、f r 、f s 、f w : eigenvectors representing train operation data, dispatching command data, track condition data, signal information data, and meteorological data respectively; S22. Use a deep neural network to encode each type of feature vector separately to form an intermediate representation h x ,h c ,h r ,h s ,h w , where DNN represents the deep neural network function, h x =DNN(f x ) represents the intermediate representation of train operation data, h c =DNN(f c ) represents the intermediate representation of the scheduling command data, h r =DNN(f r ): represents the intermediate representation of track status data, h s =DNN(f s ) represents the intermediate representation of signal information data, h w =DNN(f w ) represents an intermediate representation of meteorological data; S23. Use the self-attention mechanism to perform weighted fusion of different types of intermediate representations; S24. Based on the calculated attention weights, perform weighted summation on the intermediate representations to form the fused feature representation H: H=α x h x +α c h c +α r h r +α s h s +α w h w ; Where H represents the fused feature representation; S25, the fused feature representation H is processed through the fully connected layer to reduce the dimension and obtain a unified dataset representation {z k }, where: z k =FC(H); where z k Represents the feature representation after dimensionality reduction; S26, for a unified dataset representation {z k } is normalized to form the final fusion data set {z′ k }; S3. Extract key features from the fused dataset, use the autoencoder method for time series feature extraction, and use XGBoost for feature selection, including: S31, the fused railway dispatching safety dataset z′ k Perform preliminary processing and construct the feature matrix F = {f kj }, where f kj represents the jth feature of the kth sample, the feature is the track condition, z′ k represents the normalized kth sample, including train operation data, dispatching commands, track conditions, signal information and meteorological data, and F represents the feature matrix of all samples; S32, use the autoencoder method to extract time series features from the feature matrix F, and build an autoencoder model, including the encoder and decoder parts. The encoder takes the input feature f kj Mapped to the hidden layer representation h k , the decoder represents the hidden layer h k Reconstruct back to the input features, the optimization goal is: Among them, θ represents the parameters of the autoencoder model, including the encoder parameters θ E and decoder parameters θ D , represents the reconstructed feature matrix, Where E(θ E ,F) represents the output of the encoder, D(θ D ,E(θ E ,F)) represents the output of the decoder, || represents the norm of the reconstruction error, usually the L2 norm, λ represents the regularization parameter used to control the complexity of the model, L represents the number of layers of the autoencoder, W l represents the weight matrix of the lth layer, b l Represents the bias vector of the lth layer; S33. Extracting time series features {h k }: h k =E(θ E ,f k ); Among them, h k Indicates the historical operation mode of the train, the change of track conditions, and the change of signal status, E(θ E ,f k ) indicates that the encoder transforms the feature f of the kth sample k The process of mapping to the hidden layer; S34. Use the XGBoost algorithm for feature selection and build a set of decision trees. The objective function is: Where Θ represents the parameters of the XGBoost model, N represents the number of samples, that is, the number of samples of railway scheduling safety data, L represents the loss function, which is used to measure the error between the predicted value and the true value, and y i represents the true label of the i-th sample, such as the safety status of train scheduling, represents the predicted value of the i-th sample, T represents the number of decision trees, Ω(f t ) represents the regularization term of the t-th tree: Among them, γ represents the regularization parameter, which controls the complexity of the tree, d represents the number of features, and w j represents the weight of feature j; S35. Calculate the importance score of each feature I j ; S36, according to the importance score I j Sort the features, select the top n features with the highest scores, and construct the final feature set F′ to reflect the factors of railway scheduling safety, where n is the preset number of features; S37. Using the final feature set F′ as input data for subsequent model training to further improve the accuracy and reliability of railway scheduling safety prediction; S4. Use the processed data set to train the AI ​​model, select random forest for training, and optimize the model parameters through hyperparameter tuning technology; S5. Real-time prediction and decision support: Apply the trained AI model to real-time data to make real-time predictions of safety hazards and scheduling risks, and provide decision support and safety warnings.

2. The railway dispatching safety data aggregation processing method based on AI model training according to claim 1 is characterized in that: The S1 comprises the following steps: S11. Collecting railway dispatch-related data in real time from multiple data sources, including train operation data, dispatching commands, track conditions, signal information, and meteorological data; S12. Express the collected train operation data as a time series x t , where x t represents the train position, speed and acceleration at time t; S13, structurally process the dispatch command data and extract key information c in the dispatch command i , where c i Represents the content and timestamp of the i-th scheduling command; S14, perform spatial mapping on the track condition data to generate a track condition matrix R = {r ij }, where r ij represents the status of the track segment i at the jth moment; S15. Collect signal information data and convert it into a signal state sequence s t , where s t Represents the signal state at time t; S16, obtain meteorological data, and perform spatial and temporal matching to form a meteorological condition matrix W = {w ij }, where w ij represents the meteorological conditions at the i-th location at the j-th time; S17. Perform data cleaning on all collected data, including processing missing values, outliers, and noisy data. Use interpolation to process missing values ​​and use triple standard deviation method to detect and process outliers. S18. Standardize the cleaned data and use the z-score standardization method to convert the data into a standard normal distribution with a mean of 0 and a standard deviation of 1.

3. The railway dispatching safety data aggregation processing method based on AI model training according to claim 1 is characterized in that: The S35 specifically includes: S351, for each round of iteration t, using the current residual Build a new decision tree and update the model's predictions: in, represents the predicted value of the i-th sample after the t+1th round of iteration, represents the prediction of the railway scheduling safety data by the model after the t+1th round of iteration, η: learning rate, used to control the update amplitude of the predicted value at each iteration, f t (x i ) represents the predicted value of the t-th tree for sample i, and represents the contribution of the t-th decision tree to the railway scheduling safety data; S352. In each round of iteration, the node splitting gain of the current tree is calculated. The gain formula is: Among them, G s Represents the splitting gain of node s, which represents the information gain obtained by the current node when splitting, I L represents the sample set of the left child node, represents the railway scheduling safety data sample belonging to the left child node after splitting, I R represents the sample set of the right child node, represents the railway scheduling safety data samples belonging to the right child node after the split, I represents the sample set of the current node, represents all railway scheduling safety data samples before the split, g i Represents the first-order gradient of the i-th sample, which represents the derivative of the loss function to the predicted value and is used to measure the prediction error. i represents the second-order gradient of the i-th sample, represents the second-order derivative of the loss function with respect to the predicted value, and is used to measure the rate of change of the prediction error. λ represents the regularization parameter, which is used to control the complexity of the model and avoid overfitting. S353. For each feature j, accumulate its split gain on all nodes and calculate the importance score I j : Among them, I j represents the importance score of the jth feature, which represents the importance of feature j in all decision trees, T represents the number of decision trees, which represents the total number of decision trees constructed, and S t represents the set of all nodes in the t-th tree, represents all nodes in the t-th decision tree, G js It represents the splitting gain of the jth feature at node s, which means the information gain obtained when feature j is split at node s; S354. For each node s, the gain is calculated based on the accumulated values ​​of the first-order gradient and the second-order gradient. The calculation formula is: Among them, G js It represents the splitting gain of the jth feature at node s, and it represents the contribution of feature j when splitting at node s. s represents the sample set on node s, represents all railway scheduling safety data samples belonging to node s, g i Represents the first-order gradient of the i-th sample, which represents the derivative of the loss function to the predicted value and is used to measure the prediction error. i It represents the second-order gradient of the i-th sample and the second-order derivative of the loss function with respect to the predicted value, which is used to measure the rate of change of the prediction error.

4. The railway dispatching safety data aggregation processing method based on AI model training according to claim 1 is characterized in that: The S4 specifically includes: S41, using the processed feature set F' to construct a training data set {(x i ,y i )}, where F′ represents the final feature set, which contains the features with the highest scores. The features represent the key information of railway scheduling safety data, y i represents the label of the i-th sample, indicating the safety status of railway scheduling, such as safety, potential risk or failure, x i The feature vector representing the i-th sample contains train operation data, dispatching commands, track conditions, signal information, and meteorological data; S42, select the random forest algorithm for model training, and construct multiple decision trees {T m }, where each tree is trained using a different subset of features and samples: {T m }={T1,T2,…,T M }; Among them, {T m } represents the set of decision trees in the random forest, T m : the mth decision tree, M: the number of decision trees, indicating the total number of trees contained in the random forest; S43. Optimize model parameters through hyperparameter tuning technology to determine the best hyperparameter combination {θ k }, including the number of trees M, the maximum depth D, and the minimum number of sample splits N min : {i} k }={θ1,θ2,…,θ K }; Among them, {θ k } represents the set of hyperparameters, θ k : The kth hyperparameter, K represents the number of hyperparameters, M represents the number of decision trees, which represents the total number of trees included in the random forest, D represents the maximum depth of the decision tree, which represents the maximum number of layers of the tree, and N min Indicates the minimum number of sample splits, which indicates the minimum number of samples required for a node to split; S44. Use the grid search method to search for the best parameter combination in the preset parameter space. The optimization goal is: Among them, θ represents the hyperparameter to be optimized, N: the number of samples, which represents the total number of samples in the training data set, Loss function, which measures the predicted value and the true value y i the differences between; S45. Evaluate model performance through cross-validation; S46. Using the best hyperparameter combination Retrain the random forest model to get the final model in, The final random forest model represents the predicted value of the input feature vector x, T m (x): The predicted value of sample x by the mth decision tree. M represents the number of decision trees and the total number of trees contained in the random forest.

5. The railway dispatching safety data aggregation processing method based on AI model training according to claim 4 is characterized in that: The S45 specifically includes: S451, the training data set {(x i ,y i )} is divided into K non-overlapping subsets, denoted as {D1,D2,…,D K }, where the number of samples contained in each subset is Among them, {(x i ,y i )} represents the training dataset, which contains the features and labels of railway scheduling safety data. K represents the number of cross-validation folds, and the training dataset is divided into K subsets. k represents the kth subset, each subset contains a part of railway scheduling safety data samples, N represents the total number of samples in the training data set, and represents the number of samples of railway scheduling safety data; S452. Perform the following operations on each hyperparameter combination θ: For the kth subset D k , which is used as the validation set, and the remaining K-1 subsets {D1,D2,…,D k-1 ,D k+1 ,…,D K }Merge into training set: represents the training set in the kth fold, which is formed by merging the K-1 subsets and contains the features and labels of railway scheduling safety data; represents the validation set in the k-th fold, which consists of the k-th subset and contains the features and labels of railway scheduling safety data; In the training set Train the random forest model and get the model in, represents the predicted value of the input feature vector x obtained by the random forest model trained on the training set in the kth fold, and represents the prediction of railway scheduling safety data. represents the mth decision tree in the kth fold, the predicted value of the input feature vector x, represents the prediction of railway scheduling safety data, M represents the number of decision trees, and represents the total number of trees contained in the random forest; Using a validation set Evaluate model performance and calculate loss function The average value of: Among them, L k represents the average loss in the kth fold, represents the model performance on the validation set, represents the number of samples in the validation set, Represents the model in the kth fold for the validation set sample x i The predicted value of Represents the loss function, which measures the predicted value and the true value y i The difference between them is used to evaluate the prediction accuracy of railway dispatching safety data; S453. Calculate the average loss of all compromises as the performance indicator of the hyperparameter combination θ: Among them, L cv (θ) represents the cross-validation average loss of the hyperparameter combination θ, represents the model performance in all folds, K represents the number of cross-validation folds, L k represents the average loss in the kth fold; S454. Select the hyperparameter combination with the minimum cross-validation average loss as the best hyperparameter combination: in, represents the best hyperparameter combination after optimization, represents the kth optimized hyperparameter, L cv (θ) represents the cross-validation average loss for the hyperparameter combination θ.

6. The railway dispatching safety data aggregation processing method based on AI model training according to claim 1 is characterized in that: The S5 specifically includes: S51. Collect and pre-process real-time railway dispatching data, including train operation data, dispatching commands, track conditions, signal information and meteorological data, and convert these data into feature vectors x real-time , Among them, x real-time A feature vector representing real-time data, including the latest train status, dispatching commands, track conditions, signal information, and weather data; S52, the real-time data feature vector x real-time Input to the trained AI model Generate real-time prediction results in, Indicates the prediction result of the AI ​​model on real-time data, indicating the safety status of the current railway scheduling. Represents a trained AI model used to predict the input feature vector x; S53. Based on the prediction results Analyze safety hazards and scheduling risks, and determine whether there are potential risks based on preset thresholds: Among them, Risk represents the real-time risk level, which is classified according to the prediction results, τ high Indicates a high risk threshold, exceeding this value indicates high risk, τ low Indicates a low risk threshold, below which the risk is low; S54. Provide corresponding decision support based on the risk level, including scheduling adjustment suggestions and safety measures suggestions. The specific steps are: when the risk level is high, take emergency scheduling adjustment measures immediately; when the risk level is medium, issue an early warning and recommend that the dispatcher take preventive measures; when the risk level is low, continue monitoring without taking additional measures.

Images