Intelligent rail transit signal system and automatic scheduling method thereof

Abstract

The invention discloses an intelligent rail transit signal system and an automatic scheduling method, and relates to the technical field of rail transit, and the method comprises the steps of data acquisition, machine learning modeling, scheduling optimization algorithm based on machine learning, scheduling strategy generation, real-time signal control, automatic scheduling decision, and system monitoring and optimization. By adding the physical constraint to guide the generation process, the problem that the rail transit data generation sample does not conform to the actual physical law is solved; the problems of insufficient data generation diversity, generation mode collapse and the like are solved through methods of dynamic memory mechanism optimization generator training and the like. The problems of high-dimensional redundancy, nonlinear structural features, poor model generalization ability and the like of rail transit data are solved by using a negentropy excitation function.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit, and in particular to an intelligent rail transit signal system and an automatic dispatching method thereof. Background Art

[0002] Rail transit systems are an important part of modern urban transportation networks, and their operating efficiency and safety have a significant impact on urban development and the quality of residents' travel. However, with the acceleration of urbanization, rail transit faces increasingly severe challenges, including the rapid growth of passenger flow, the complexity of the operating environment, and the frequent occurrence of emergencies. These problems have put forward higher requirements for the automatic dispatching and signal control of rail transit systems.

[0003] The existing methods for automatic scheduling of rail transit have problems such as data expansion methods failing to effectively consider the physical constraints of rail transit data, resulting in poor quality of generated rail transit data, imbalanced categories in the distribution of rail transit data, traditional dimensionality reduction methods failing to effectively handle the high-dimensional redundancy of rail transit data, failing to capture important topological features in rail transit data, dimensionality reduction models being susceptible to local optimal problems, and difficulty in achieving stable performance in complex rail transit scenarios. Summary of the invention

[0004] In order to solve the technical problems existing in the automatic dispatching of rail transit in the prior art, the present invention provides an intelligent rail transit signal system and an automatic dispatching method thereof.

[0005] The present invention is achieved through the following technical solutions: An intelligent rail transit signal automatic dispatching method, comprising: S1: Data collection: real-time collection of rail transit operation data, labeling and preprocessing of data; S2: Machine learning modeling, using historical data and real-time data to build a prediction model, and generate a scheduling strategy based on a scheduling optimization algorithm based on machine learning; S3: Real-time signal control, dynamically adjusting signal status according to system decisions; S4: Automatic dispatch decision-making, generating an optimized dispatch plan for trains based on integrated system information and machine learning modeling results; S5: System monitoring and optimization, to monitor the operating status of the entire rail transit system in real time and continuously optimize system performance based on feedback.

[0006] Furthermore, the step S1 also includes performing small sample identification on the collected data set by using statistical analysis and machine learning algorithms. If a class imbalance or insufficient sample problem is detected, a generative adversarial network algorithm based on an augmentation strategy is used as a sample expansion model to expand rail transit data; The training process of the generative adversarial network algorithm based on the augmentation strategy includes initializing the standard structure of the generative adversarial network, including the network architecture of the generator and the discriminator; the generator is used to generate samples close to the distribution of real rail transit data, and the discriminator is used to determine whether the rail transit data is a real sample or a false sample generated by the generator.

[0007] Furthermore, during the training process of the generative adversarial network, the physical constraint function is also added to the loss function of the generator.

[0008] Furthermore, during the training process of the generative adversarial network, a dynamic memory mechanism is also used to store and update effective sample features and output information of the generator during the historical training process.

[0009] Furthermore, the game relationship between the generator and the discriminator is dynamically balanced through an adaptive step size adjustment mechanism, and the learning rates of the generator and the discriminator are dynamically adjusted according to the current training state and changes in the target distribution.

[0010] Furthermore, step S2 also includes: using an autoencoder based on adaptive gradient correction as a feature dimensionality reduction model, and optimizing the autoencoder network by using a dynamic adjustment mechanism based on data topology analysis and an adaptive gradient correction method.

[0011] Furthermore, an excitation function based on negative entropy driving is adopted for the output of the coding layer.

[0012] Furthermore, the entropy weights of rail transit data features are dynamically adjusted, and a method based on topological rail transit data analysis is used to capture the nonlinear structural characteristics of rail transit data.

[0013] The present invention also provides an intelligent rail transit system, based on the automatic dispatching method as described above, which comprises: The intelligent rail transit signal system includes a data acquisition unit, a machine learning modeling unit, a real-time signal control unit, an automatic scheduling decision unit, and a system monitoring and optimization unit; The data acquisition unit is used to collect key data in rail transit operation in real time, and to mark and expand it; The machine learning modeling unit is used to establish a prediction model using historical data and real-time data, develop a scheduling optimization algorithm based on machine learning, and generate a scheduling strategy; The real-time signal control unit is used to receive predictions and instructions provided by the machine learning modeling unit and the automatic scheduling decision unit, and dynamically adjust the signal state according to the system decision; The automatic dispatch decision unit is used to generate an optimized dispatch plan for trains by integrating system information and machine learning modeling results; The system monitoring and optimization unit is used to monitor the operating status of the entire rail transit system in real time and continuously optimize system performance based on feedback.

[0014] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, on which program instructions of an intelligent rail transit signal automatic scheduling method are stored. The program instructions of the intelligent rail transit signal automatic scheduling method can be executed by one or more processors to implement the steps of the intelligent rail transit signal automatic scheduling method as described above.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention solves the problem that rail transit data generation samples do not conform to actual physical laws by adding physical constraints to guide the generation process.

[0016] (2) The present invention performs data expansion and optimizes generator training through a dynamic memory mechanism to solve the problem of insufficient data generation diversity. By adjusting the adaptive learning rate, the adversarial training of the generator and the discriminator is balanced to avoid the problem of generation mode collapse.

[0017] (3) The present invention adopts an autoencoder based on adaptive gradient correction and uses a negative entropy excitation function to enhance the preference for high-value features, thereby solving the high-dimensional redundancy problem of rail transit data. It combines topological data analysis methods to capture the nonlinear structural characteristics in rail transit data, thereby solving the problem that traditional dimensionality reduction methods are difficult to retain key topological information. It optimizes the dimensionality reduction process through a dynamic entropy weight adjustment mechanism, thereby solving the problem of poor model generalization ability caused by data distribution heterogeneity. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 It is a schematic diagram of the process flow of the intelligent rail transit signal automatic scheduling method according to an embodiment of the present application; Figure 2 is a correlation analysis diagram between performance and physical constraint strength according to an embodiment of the present application; Figure 3 is a comparison chart of data diversity between the augmented algorithm according to the embodiment of the present application and the conventional generative adversarial network algorithm under different training rounds; Figure 4 is a comparison chart of the decreasing trend of the loss value during the training process between the adaptive gradient correction autoencoder according to the embodiment of the present application and the conventional method; Figure 5It is the distribution of original data according to the embodiment of the present application, conventional technology reconstructed data and adaptive gradient correction autoencoder reconstructed data. DETAILED DESCRIPTION

[0019] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0020] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0021] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention. The illustrations only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the form, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0022] See also Figure 1 , an intelligent rail transit signal automatic dispatching method, comprising the following steps:

[0023] S1: Data collection, including:

[0024] S11: Collect rail transit operation data in real time and use it as a training data set for machine learning models after annotation;

[0025] The data collected comes from various sensors in the rail transit operation environment, including but not limited to track sensors, camera monitoring equipment, on-board equipment and infrastructure monitoring equipment;

[0026] The data is collected through real-time dynamic capture to ensure that the collected information can reflect the instant status of the train and the real-time changes of the track environment;

[0027] The collected data is first formatted by the local processing node, and then stored in a high-performance data server in a structured data format. The main storage format is a time series database, which is convenient for subsequent real-time query and historical data analysis.

[0028] In one embodiment, the collected data includes:

[0029] Location Information , record the specific position of the train on the track; speed , the real-time running speed of the train; acceleration , real-time acceleration of the train; train number information , train number; station stop time , the train’s stop time at each station; signal status , status information of the signal lights on the track; number of passengers , real-time passenger data; energy consumption data , energy consumption during train operation; equipment status , operating status of on-board and track equipment; environmental factors , such as temperature, humidity and other environmental parameters.

[0030] It should be noted that this embodiment is only used to illustrate one data format and type of the present invention. In actual applications, the attributes of data are usually more than 10 attributes, and the number of attributes of data may reach dozens or even hundreds.

[0031] S12: labeling the collected data, optionally, the labeling method is manual labeling;

[0032] In one embodiment, the marked categories include:

[0033] 0-Normal operation: The train runs normally according to the schedule;

[0034] 1- Delay: The train runs beyond the scheduled timetable;

[0035] 2- Emergency stop: The train stops due to a malfunction or emergency;

[0036] 3-Maintenance: The train undergoes routine or unplanned maintenance;

[0037] 4- Abnormal behavior: The train behaves in an abnormal operating mode, such as abnormal speed.

[0038] S13: Data preprocessing: cleaning and filtering the collected data, removing outliers and noise, and providing accurate data input for the machine learning modeling unit.

[0039] S14: Small sample data identification: Small sample data identification is performed on the collected data set by using statistical analysis and machine learning algorithms. For example, distribution detection algorithms (such as those based on Kullback-Leibler divergence or distribution deviation indicators) are used to evaluate whether there are problems of class imbalance or insufficient samples in the data distribution.

[0040] Since the acquisition, labeling and preprocessing of training data are time-consuming and labor-intensive, and insufficient training samples can easily lead to poor generalization ability of the model, while affecting the accuracy of the model; the small sample data recognition function detects the small sample data problem that may exist in the system, and ensures the adequacy and reliability of the model training data through data expansion technology;

[0041] S15: If class imbalance or insufficient samples are detected, the generative adversarial network algorithm based on the augmentation strategy is used as the sample augmentation model to expand the rail transit data. The diversity and stability of the generated rail transit data are enhanced through adaptive data augmentation strategies and discriminator quality control methods. At the same time, an adaptive learning rate adjustment mechanism is used to balance the game between the generator and the discriminator, thereby improving training efficiency and stability.

[0042] Specifically, the training process of the generative adversarial network algorithm based on the augmentation strategy is as follows:

[0043] (1) Initialize the standard structure of the generative adversarial network, including the network architecture of the generator and the discriminator; the task of the generator is to generate samples that are as close as possible to the distribution of real rail transit data, and the discriminator is responsible for judging whether the rail transit data is a real sample or a false sample generated by the generator. The way the generator generates rail transit data is expressed as:

[0044]

[0045] In the formula, For the generated samples, is the generator function; is the input noise vector, representing randomness; is the parameter of the generator, representing the weight of the generator. Preferably, Set to the initial random value.

[0046] Furthermore, the loss function of the generator is defined as:

[0047]

[0048] In the formula, is the loss function of the generator, For the expected operation; is a noise vector, representing the input of the generator; Indicates compliance with a specific distribution; is the distribution of the noise vector, characterizing the randomness of the input; is the weight hyperparameter of the physical constraint, which controls the impact of the physical constraint on the loss function; is a physical constraint function, which indicates whether the generated rail transit data complies with the physical constraints; is the discriminator function. Preferably, Set to 0.2.

[0049] like Figure 2 The figure shows the correlation analysis results between performance and physical constraint strength verified by experiments.

[0050] (2) Add the physical constraint function to the generator’s loss function;

[0051] Rail transit data generation is not just a pure random noise transformation, but also needs to take into account the physical laws of rail transit data in the real world. The generated rail transit data should comply with certain physical restrictions or constraints. The present invention adopts a physical scene perception method, which not only focuses on the quality of generated rail transit data, but also ensures that the generated rail transit data complies with the physical laws or prior knowledge in a specific field. The training of the generation network is guided by the physical constraint term, and the physical constraint function is transformed into a physical scene perception method. Added to the loss function of the generator, it ensures that the generated samples are not only similar to the real rail transit data, but also meet the constraints of physical laws. By adopting physical constraints, the generated samples are restricted in accordance with the physical laws in the field of rail transit. For example, when expanding rail transit flow data, it ensures that the generated samples conform to the smooth distribution law of actual flow over time or space. The calculation method of the physical constraint function is expressed as:

[0052]

[0053] In the formula, To generate the Laplacian operator of rail transit data, it represents the spatial smoothness or rate of change of rail transit data; is a predefined physical constant or constraint value. Preferably, Defined as the variance value of the real rail transit data input to the generative adversarial network for the current batch.

[0054] (3) During the training process of the generative adversarial network, a dynamic memory mechanism is used to store and update the effective sample features and output information of the generator during the historical training process;

[0055] The dynamic memory mechanism adaptively selects and updates historical rail transit data according to the loss function and the quality of generated samples in the training process, provides high-quality sample memory for the training of subsequent generators, and ensures that the generator continuously optimizes the quality of the rail transit data it generates. The dynamic memory mechanism is implemented through a memory module. For example, when processing passenger flow data of different stations, the dynamic memory mechanism can help the generator more effectively capture and reproduce the characteristic differences between different stations, avoiding the generation of single or repeated data. The update and implementation of the memory module are expressed as follows:

[0056]

[0057] In the formula, For the The sample features of the iterative memory represent the memory module in the The status of the iteration; For the The sample features of the iterative memory represent the memory module in the The status of the iteration; is the updated weight coefficient, representing the balance factor of memory update; A weighting factor based on the quality of the generated sample characterizes the importance of the sample. Preferably, Set to 0.3.

[0058] Furthermore, the weighted factor based on the quality of the generated samples enables the memory mechanism to dynamically retain the generated samples that are closest to the distribution of the real rail transit data, thereby improving the training quality of the generator. The calculation method is expressed as:

[0059]

[0060] In the formula, is the L2 norm; It is real rail transit data, representing real samples; The total number of samples generated for the current batch; Indicates Real rail transit data; is a smoothing factor to avoid tiny constants with zero denominators; is the feature weight matrix of the sample. Preferably, Set to .

[0061] Furthermore, in order to achieve adaptive non-local feature interaction, non-local operations are used to enhance global feature interaction. The relationship between different locations of the generated samples is represented by the feature weight matrix, which helps to capture the long-range dependencies in rail transit data, such as the connection between different stations or the correlation of passenger flow in different time periods, so that the generated samples are more in line with the characteristics of the actual rail transit network, especially when simulating the coordinated flow changes of multiple stations. The calculation method is expressed as:

[0062]

[0063]

[0064] In the formula, Representation characteristics and The dot product between them represents the similarity between features; To generate the rail transit data feature vector The eigenvalues ​​of the positions; To generate the rail transit data feature vector The eigenvalues ​​of the positions; To generate the characteristic dimension of the sample; To generate the sample The feature relative to The weight of a feature.

[0065] (4) Conduct adversarial training of the generative adversarial network. The generator randomly generates rail transit data samples, and the discriminator judges their authenticity. The discriminator adjusts parameters according to feedback to make the rail transit data generated by the generator closer to the real distribution. Different from the traditional generative adversarial network, the present invention sets the generator to combine the historical sample features stored in the dynamic memory mechanism during each training, and adjusts the generation strategy according to the past training information to avoid falling into the local optimal solution, thereby improving the diversity and quality of the generated rail transit data. The discriminator loss function of the generative adversarial network is expressed as:

[0066]

[0067] In the formula, is the loss function of the discriminator; It is the distribution of real rail transit data and represents the real distribution of rail transit data.

[0068] (5) During the training process, an adaptive data augmentation strategy is used to transform and enhance each batch of generated samples, including rotation, scaling, translation and other operations, to simulate the diversity and complexity of rail transit data distribution. The augmentation strategy is dynamically adjusted according to the feedback of the discriminator, focusing on strengthening the performance of generated rail transit data in specific modes to ensure that the generator learns the diversity and regularity of rail transit data. For example, for the fluctuation of passenger flow during peak hours in the morning and evening, more samples during peak hours are generated through augmentation operations to improve the balance of training data distribution. The implementation method of the augmentation strategy is expressed as:

[0069]

[0070] In the formula, is the augmentation strategy factor, which represents the influence of the data augmentation operation; is the augmentation control factor, which represents the weight of the discriminator feedback in the augmentation strategy; is the gradient of the augmentation operation, representing the adjustment direction of the augmentation strategy according to the discriminator feedback. Preferably, Set to 0.5.

[0071] Furthermore, the gradient of the augmentation operation is represented by the derivative of the difference between the generator output and the real rail traffic data, and the calculation method is expressed as:

[0072]

[0073] In the formula, is the augmented sample, representing the rail transit data after augmentation operation; is the symbol of partial derivative.

[0074] Furthermore, the augmentation operation includes transformations such as translation, rotation, and scaling, which act on the generated samples through the transformation function and generate new samples, which can be expressed as:

[0075]

[0076] Where; is the transformation function, representing the specific augmentation operation; is the augmentation strategy factor, is the sample before augmentation.

[0077] In one embodiment, the implementation of the transformation function includes the following specific transformations:

[0078] 1- Translation operation, expressed as:

[0079]

[0080] In the formula, is the translation operation function, which represents the translation transformation of rail transit data; is a translation vector, representing the magnitude and direction of the translation. Preferably, Set to a fixed value, such as 3.

[0081] 2-Rotation operation, expressed as:

[0082]

[0083] In the formula, is the rotation operation function, which represents the rotation transformation of rail transit data; is the rotation matrix, representing the angle of rotation; is the rotation angle, representing the magnitude of the rotation. Preferably, Set to 30 degrees.

[0084] 3- Scaling operation, expressed as:

[0085]

[0086] In the formula, is the scaling operation function, which represents the scaling transformation of rail transit data; is a scaling factor, representing the scaling ratio. Preferably, Set to 1.2.

[0087] (6) During the training process, the discriminator optimizes its classification ability by combining the similarity information between the generated rail transit data and the historical rail transit data, distinguishing the real rail transit data from the generated rail transit data. At the same time, combined with the sensitivity of the generator to the transformed samples, the fine-grained feature information of the generated rail transit data is added to the feedback of the discriminator to help the discriminator better adapt to the changes of the generator, such as ensuring that the generated passenger flow data is consistent with the changing trend of the historical data. In this process, in order to maintain the stability of the training process, the loss function of the discriminator adopts a stability adjustment mechanism based on the quality of the generated rail transit data, which is expressed as:

[0088]

[0089] In the formula, is the discriminator loss function after stability adjustment; is the stability adjustment factor, which represents the weight of quality control; is the quality control loss of the discriminator. Preferably, Set to 0.1.

[0090] Furthermore, the quality control loss of the discriminator is combined with the regularization term of the rail transit data quality, which means that the discriminator not only judges the generated samples, but also needs to evaluate the quality of the samples. The calculation method is expressed as:

[0091]

[0092] In the formula, is the quality control regularization coefficient, which represents the weight of the regularization term. Preferably, Set to 0.1.

[0093] (7) During the training process, the game relationship between the generator and the discriminator is dynamically balanced through an adaptive step size adjustment mechanism. The learning rates of the generator and the discriminator are dynamically adjusted according to the current training state and the change of the target distribution to avoid excessive adjustment of the generator or too strong a discriminator, thereby improving the stability and efficiency of the training. In the rail transit data expansion, balanced training is used to reduce the mode collapse caused by improper parameter setting and ensure the diversity of generated samples. The adaptive learning rate adjustment method is expressed as:

[0094]

[0095] In the formula, is the learning rate of the generator and the discriminator, representing the step size of parameter update; is the initial learning rate of the generator and discriminator, representing the starting step of training; is the learning rate adjustment factor of the generator and the discriminator, representing the sensitivity of the step size adjustment. Preferably, Set to 0.1.

[0096] (8) Update the parameters of the generator and the discriminator according to the gradient descent method, and repeat the above steps until the preset stop iteration condition is met, indicating that the model training is completed. In one embodiment, the preset stop iteration condition is to reach a preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times.

[0097] After the rail transit data expansion model training is completed, the trained rail transit data expansion model is used to increase the number of samples. In one embodiment, assuming that the original collected samples are 800, and the rail transit data expansion model expands and generates 200 samples, the expanded rail transit data set contains 1000 samples.

[0098] like Figure 3 Shown is a comparison of data diversity between the augmented algorithm and the conventional generative adversarial network algorithm at different training rounds.

[0099] S2: Machine learning modeling; providing prediction models and optimization algorithm support for the system's intelligent decision-making, including the following steps:

[0100] S21: Use historical data and real-time data to establish prediction models, including train running time prediction, passenger flow prediction, and the impact of emergencies on traffic;

[0101] S22: Develop scheduling optimization algorithms based on machine learning to generate scheduling strategies that meet actual needs;

[0102] These include: using an autoencoder machine learning model to reduce the dimension of the expanded training data set, and using a preset Softmax function to classify the reduced feature vectors to obtain classification prediction categories.

[0103] In order to solve the problems of information redundancy and the model being prone to falling into local optimum in the process of dimensionality reduction of high-dimensional rail transit data, the present invention adopts an autoencoder based on adaptive gradient correction as a feature dimensionality reduction model, and utilizes a dynamic adjustment mechanism based on data topology analysis and an adaptive gradient correction method to optimize the performance of the autoencoder network in complex rail transit data distribution and high-dimensional feature space, thereby achieving more efficient feature compression and reconstruction.

[0104] Specifically, the training process of the autoencoder algorithm based on adaptive gradient correction is as follows:

[0105] a. Initialize the weights and biases of the autoencoder network randomly, and the initialized parameters obey a normal distribution with a mean of 0 and a variance of the identity matrix.

[0106] b. Compress the input rail transit data features through the encoder and convert them into low-dimensional coding representation to obtain the main information of the rail transit data and ignore redundancy, which helps to deal with the common high-dimensional redundancy problems in rail transit data, such as compressing complex trajectory data into key path features to reduce computational overhead. The calculation method of the encoding output is expressed as:

[0107]

[0108] In the formula, is the output of the encoding layer, is the activation function of the encoder, is the weight of the encoder, is the input rail transit data of the encoder, Preferably, the activation function of the encoder adopts the Sigmoid activation function.

[0109] c. The output of the coding layer uses an activation function based on negative entropy drive, which uses negative entropy to measure the information content in the features, thereby improving the model's preference for key features. By measuring the information content and focusing on high-value features, the model can more effectively select important variables that affect the overall operating efficiency in rail transit data. For example, the flow anomaly points are extracted from the flow matrix between stations. The calculation method is expressed as:

[0110]

[0111] In the formula, is negative entropy, is the first elements, is the first The probability density of an element.

[0112] d. The encoded rail transit data is restored to its original dimension through the decoder, and the difference between the original rail transit data and the reconstructed rail transit data is compared to guide the learning of the autoencoder network. The calculation method is expressed as:

[0113]

[0114] In the formula, To reconstruct rail transit data, is the weight of the decoder, is the decoder bias, is the activation function of the decoder. Preferably, the activation function of the decoder adopts the ReLU activation function.

[0115] e. Dynamically adjust the entropy weight of rail transit data features, and use a method based on topological rail transit data analysis to capture the nonlinear structural characteristics of rail transit data. It can adaptively identify and retain important topological information in high-dimensional rail transit data, such as maintaining important topological information of core stations in passenger flow prediction, thereby effectively improving the performance in classification tasks after dimensionality reduction. The calculation method is expressed as:

[0116]

[0117] In the formula, For the The entropy weight of the samples, is the basic adjustment coefficient, is the penalty degree coefficient, is the feature dimension after dimensionality reduction, is the coefficient of topological feature influence; For the Sample No. The topological influence of each feature is calculated by the topological rail transit data analysis method. The topological rail transit data analysis (TDA) is a method that combines algebraic topology and computational geometry to extract geometric and topological features from complex rail transit data sets. Its main goal is to analyze the shape and structure of rail transit data to reveal potential patterns and relationships; For the The weight factor of each feature is a training parameter, which is determined by gradient descent training; For the The sample The entropy value of the feature. Preferably, Set to 0.2, Set to 0.1, Set to 0.2.

[0118] f. Use the loss function to constrain the training process, and use an adaptive gradient correction algorithm in the autoencoder network update phase to alleviate local optimality and overfitting in high-dimensional training, and improve the search efficiency of the model in complex rail transit data scenarios (such as complex path classification tasks in optimizing trajectory pattern recognition). The calculation method of the loss function is expressed as:

[0119]

[0120] In the formula, is the overall loss function of the autoencoder network, is the L2 norm, For the The entropy value of samples is The regularization coefficient of the autoencoder, is the balance coefficient of the topological feature in the loss function. Preferably, Set to 0.3, Set to 0.3.

[0121] In one embodiment, by comparing the decreasing trend of the loss value during the training process, the superiority of the adaptive gradient correction autoencoder in training speed and convergence ability is demonstrated, such as Figure 4 shown.

[0122] g. Adopt adaptive gradient correction to update the parameters of the autoencoder network to improve the convergence speed and robustness of the model under complex high-dimensional rail transit data. The gradient correction and parameter update method are expressed as follows:

[0123]

[0124] In the formula, is the initial gradient of the autoencoder network parameters, are the parameters of the autoencoder, including the weights of the encoder and the decoder.

[0125] Furthermore, the initial gradient is corrected according to the adaptive adjustment strategy, and the calculation method is expressed as:

[0126]

[0127] In the formula, is the gradient of the modified autoencoder network parameters, is the learning rate corrected by the autoencoder gradient, is the gradient adjustment coefficient; is a sign function, representing the maintenance of gradient direction consistency. Preferably, Set to 0.3, Set to 0.1.

[0128] Furthermore, the network parameters of the autoencoder are updated based on the corrected gradient, and the calculation method is expressed as:

[0129]

[0130] In the formula, are the updated autoencoder network parameters, are the autoencoder network parameters before updating.

[0131] h. Repeat the above steps until the preset stop iteration condition is met, indicating that the model training is completed. In one embodiment, the preset stop iteration condition is reaching a preset maximum number of iterations. Preferably, the preset maximum number of iterations is set to 1000 times.

[0132] In one embodiment, the data is reduced to 2D by principal component analysis, the distribution of the original data and the reconstructed data is visualized, and the performance of different methods on data reconstruction is compared. The results are as follows: Figure 5 As shown, feature 1 and feature 2 are the features of projecting data into two-dimensional space through principal component analysis, wherein feature 1 is the first principal component and feature 2 is the second principal component.

[0133] After the autoencoder model training is completed, the feature vector after dimension reduction is classified using the preset Softmax function to obtain the classification prediction category. In one embodiment, the categories include:

[0134] 0-Normal operation: The train runs normally according to the schedule;

[0135] 1- Delay: The train runs beyond the scheduled timetable;

[0136] 2- Emergency stop: The train stops due to a malfunction or emergency;

[0137] 3-Maintenance: The train undergoes routine or unplanned maintenance;

[0138] 4- Abnormal behavior: The train behaves in an abnormal operating mode, such as abnormal speed.

[0139] S23: Dynamically adjust model parameters and update the model in real time based on the latest data to improve the accuracy and robustness of the prediction.

[0140] S24: Provide data visualization tools for analyzing modeling results and predicting performance, and provide data support for other units.

[0141] S3: Real-time signal control; specifically, dynamically adjust signal status according to system decisions to ensure the safety and efficiency of train operation.

[0142] Real-time signal control features include:

[0143] Receive predictions and instructions from the machine learning modeling unit and the automatic scheduling decision unit, and adjust the status of traffic lights and track switching positions according to the optimization results.

[0144] Adaptive signal control is achieved to support priority handling in emergency situations, such as signal priority changes in the event of train delays or emergencies.

[0145] Two-way information exchange is achieved with the train through the communication system to ensure the consistency of signal control and train operation.

[0146] S4: Automatic dispatch decision, generating an optimized dispatch plan for trains based on comprehensive system information and machine learning modeling results.

[0147] Automatic scheduling decision functions include:

[0148] Based on passenger flow forecasts, operating status and line conditions, the train timetable is automatically planned, and the departure intervals and stop times are dynamically adjusted.

[0149] Quickly generate emergency dispatch plans in the event of emergencies (such as equipment failure, severe weather or unexpected delays) to minimize the impact on the operation of the entire network.

[0150] Intelligent allocation of resource utilization, such as track use rights and platform allocation, ensures maximum line operation efficiency.

[0151] S5: System monitoring and optimization, to monitor the operating status of the entire rail transit system in real time and continuously optimize system performance based on feedback.

[0152] The functions of the system monitoring and optimization unit include:

[0153] The train operation status, signal status and key system performance indicators are displayed in real time, providing a comprehensive monitoring view of the system status.

[0154] Record and analyze operation logs to identify potential problems, such as reduced operating efficiency or signal control delays, and provide a basis for subsequent optimization.

[0155] Regularly evaluate the performance of machine learning models and scheduling algorithms, generate performance reports and provide feedback to the modeling unit for improvement.

[0156] Supports multi-level early warning mechanism to identify potential operational risks in advance and issue early warning information so that the system can be adjusted in time.

[0157] In this embodiment, the present invention solves the problem that the rail transit data generation samples do not conform to the actual physical laws by adding physical constraints to guide the generation process in the data expansion method based on the generative adversarial network; optimizes the generator training through the dynamic memory mechanism to solve the problem of insufficient data generation diversity; and balances the adversarial training of the generator and the discriminator through adaptive learning rate adjustment to avoid the problem of generation mode collapse. In the autoencoder based on adaptive gradient correction, the negative entropy excitation function is used to enhance the preference for high-value features to solve the high-dimensional redundancy problem of rail transit data; combines the topological data analysis method to capture the nonlinear structural characteristics in the rail transit data to solve the problem that the traditional dimensionality reduction method is difficult to retain key topological information; optimizes the dimensionality reduction process through the dynamic entropy weight adjustment mechanism to solve the problem of poor model generalization ability caused by data distribution heterogeneity.

[0158] The embodiment of the present invention further provides an intelligent rail transit signal system, which includes a data acquisition unit, a machine learning modeling unit, a real-time signal control unit, an automatic scheduling decision unit, and a system monitoring and optimization unit;

[0159] The data acquisition unit is used to collect key data in rail transit operation in real time, and after annotation, it is used as a training data set for the machine learning model; it also includes data preprocessing and small sample data identification. If the problem of class imbalance or insufficient samples is detected, the generative adversarial network algorithm based on the augmentation strategy is used as a sample expansion model to expand rail transit data;

[0160] The machine learning modeling unit is used to establish a prediction model using historical data and real-time data, develop a scheduling optimization algorithm based on machine learning, and generate a scheduling strategy;

[0161] The real-time signal control unit is used to receive predictions and instructions provided by the machine learning modeling unit and the automatic scheduling decision unit, and dynamically adjust the signal state according to the system decision;

[0162] The automatic dispatch decision unit is used to generate an optimized dispatch plan for trains by integrating system information and machine learning modeling results;

[0163] The system monitoring and optimization unit is used to monitor the operating status of the entire rail transit system in real time and continuously optimize system performance based on feedback.

[0164] In addition, an embodiment of the present invention also proposes a computer-readable storage medium, on which program instructions of an intelligent rail transit signal automatic scheduling method are stored. The program instructions of the intelligent rail transit signal automatic scheduling method can be executed by one or more processors to implement the steps of the intelligent rail transit signal automatic scheduling method as described above.

[0165] The embodiments described above are only descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. An intelligent rail transit signal automatic dispatching method, characterized in that: include: S1: Data collection: real-time collection of rail transit operation data, labeling and preprocessing of data; S2: Machine learning modeling, using historical data and real-time data to build a prediction model, and generate a scheduling strategy based on a scheduling optimization algorithm based on machine learning; S3: Real-time signal control, dynamically adjusting signal status according to system decisions; S4: Automatic dispatch decision-making, generating an optimized dispatch plan for trains based on integrated system information and machine learning modeling results; S5: System monitoring and optimization, to monitor the operating status of the entire rail transit system in real time and continuously optimize system performance based on feedback.

2. The intelligent rail transit signal automatic dispatching method according to claim 1 is characterized in that: The step S1 also includes performing small sample identification on the collected data set by using statistical analysis and machine learning algorithms. If a class imbalance or insufficient sample is detected, a generative adversarial network algorithm based on an augmentation strategy is used as a sample expansion model to expand rail transit data; The training process of the generative adversarial network algorithm based on the augmentation strategy includes initializing the standard structure of the generative adversarial network, including the network architecture of the generator and the discriminator; the generator is used to generate samples close to the distribution of real rail transit data, and the discriminator is used to determine whether the rail transit data is a real sample or a false sample generated by the generator.

3. The intelligent rail transit signal automatic dispatching method according to claim 2 is characterized in that: In the training process of the generative adversarial network, the physical constraint function is also added to the loss function of the generator; the loss function of the generator is defined as: In the formula, is the loss function of the generator, For the expected operation; is a noise vector, representing the input of the generator; Indicates compliance with a specific distribution; is the distribution of the noise vector, characterizing the randomness of the input; is the weight hyperparameter of the physical constraint, which controls the impact of the physical constraint on the loss function; is a physical constraint function, which indicates whether the generated rail transit data complies with the physical constraints; is the discriminator function; The calculation method of the physical constraint function is expressed as: In the formula, To generate the Laplacian operator of rail transit data, it represents the spatial smoothness or rate of change of rail transit data; are predefined physical constants or constraint values.

4. The intelligent rail transit signal automatic dispatching method according to claim 2 is characterized in that: In the training process of the generative adversarial network, a dynamic memory mechanism is also used to store and update the effective sample features and output information of the generator in the historical training process; the update and implementation method of the memory module is expressed as: In the formula, For the The sample features of the iterative memory represent the memory module in the The status of the iteration; For the The sample features of the iterative memory represent the memory module in the The status of the iteration; is the updated weight coefficient, representing the balance factor of memory update; A weighting factor based on the quality of the generated sample, characterizing the importance of the sample.

5. The intelligent rail transit signal automatic dispatching method according to claim 2 is characterized in that: The game relationship between the generator and the discriminator is dynamically balanced through an adaptive step size adjustment mechanism. The learning rates of the generator and the discriminator are dynamically adjusted according to the current training state and the changes in the target distribution. In the rail transit data expansion, balanced training is used to reduce the mode collapse caused by improper parameter settings and ensure the diversity of generated samples. The adaptive learning rate adjustment method is expressed as: In the formula, is the learning rate of the generator and discriminator, representing the step size of parameter update; is the initial learning rate of the generator and discriminator, representing the starting step of training; is the learning rate adjustment factor for the generator and discriminator, representing the sensitivity of step size adjustment.

6. The intelligent rail transit signal automatic dispatching method according to claim 1 is characterized in that: The step S2 further includes: using an autoencoder based on adaptive gradient correction as a feature dimension reduction model, optimizing the autoencoder network by using a dynamic adjustment mechanism based on data topology analysis and an adaptive gradient correction method; the calculation method of the encoding output is expressed as: In the formula, is the output of the encoding layer, is the activation function of the encoder, is the weight of the encoder, is the input rail transit data of the encoder, The encoder offset.

7. The intelligent rail transit signal automatic dispatching method according to claim 6 is characterized in that: The output of the coding layer adopts an excitation function based on negative entropy drive.

8. The intelligent rail transit signal automatic dispatching method according to claim 7 is characterized in that: The entropy weights of rail transit data features are dynamically adjusted, and a method based on topological rail transit data analysis is used to capture the nonlinear structural characteristics of rail transit data.

9. An intelligent rail transit signal system, based on the intelligent rail transit signal automatic dispatching method according to any one of claims 1 to 8, comprising: The intelligent rail transit signal system includes a data acquisition unit, a machine learning modeling unit, a real-time signal control unit, an automatic scheduling decision unit, and a system monitoring and optimization unit; The data acquisition unit is used to collect key data in rail transit operation in real time, and to mark and expand it; The machine learning modeling unit is used to establish a prediction model using historical data and real-time data, develop a scheduling optimization algorithm based on machine learning, and generate a scheduling strategy; The real-time signal control unit is used to receive predictions and instructions provided by the machine learning modeling unit and the automatic scheduling decision unit, and dynamically adjust the signal state according to the system decision; The automatic dispatch decision unit is used to generate an optimized dispatch plan for trains by integrating system information and machine learning modeling results; The system monitoring and optimization unit is used to monitor the operating status of the entire rail transit system in real time and continuously optimize system performance based on feedback.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions of an intelligent rail transit signal automatic scheduling method, and the program instructions of the intelligent rail transit signal automatic scheduling method can be executed by one or more processors to implement the steps of the intelligent rail transit signal automatic scheduling method as described in any one of claims 1 to 8.

Images