CN-116127395-B - Real-time abnormality sensing method for automatic protection system of high-speed train

CN116127395BCN 116127395 BCN116127395 BCN 116127395BCN-116127395-B

Abstract

The invention discloses a real-time abnormality sensing method of an automatic protection system of a high-speed train, which takes real-time operation data of all sub-equipment of an ATP (adenosine triphosphate) as an analysis object, realizes memory of a log sequence of the ATP system and correlation analysis before and after log information through a coding and decoding network, introduces a multi-head attention mechanism to process the operation data of each equipment in a layering way, and solves the concurrency requirement of data generated by multiple equipment simultaneously. And finally, systematically aggregating, accurately capturing the abnormal behavior of any piece of sub-equipment in real time, and comprehensively and timely judging the definite fault part on the minimum granularity.

Inventors

KANG RENWEI
LIU LEI
WANG FEI
CHEN HUIYUAN
CHENG JIANFENG
LI YINAN
DAI BO
LI KE
WANG YU
YI PEIRAN
SUN WENZHE
YUE LIN

Assignees

中国铁道科学研究院集团有限公司通信信号研究所
中国铁道科学研究院集团有限公司
北京华铁信息技术有限公司
北京锐驰国铁智能运输系统工程技术有限公司

Dates

Publication Date: 20260505
Application Date: 20230112

Claims (9)

1. The real-time abnormality sensing method for the automatic protection system of the high-speed train is characterized by comprising the following steps of: Constructing an abnormal perception model of an ATP system, which comprises a coding network, a decoding network, an attention mechanism layer and a classifier, wherein the ATP system is an automatic protection system of a high-speed train; The training stage comprises the steps of encoding log data of all sub-equipment in an ATP system at historical time through an encoding network, carrying out linear circulation information transmission by adopting an internal state during encoding at each time, memorizing information at the previous time, calculating hidden states by combining the internal states to obtain hidden states at all times, wherein the encoding network is realized through a long short-time memory neural network, each time, calculating candidate states, forgetting gates, input gates and output gates according to the input log data, calculating the internal states by using the forgetting gates, the input gates and the candidate states, calculating the hidden states by using the internal states and the output gates, and recording the hidden states at all times as Wherein the input log data is a log sequence formed by log data generated by a plurality of sub-devices in the ATP system, m is the length of the input log sequence, which is equal to the total time, The hidden state at the moment i is indicated, i=1.. m; the hidden state at the last moment is used as the hidden state at the initial moment of the decoding network, and then the hidden state at each moment of the decoding network is determined by combining the hidden state at the last moment with the hidden state of the coding network at all moments through a multi-head attention mechanism layer, and the calculated attention function is determined; And in the sensing stage, abnormal sensing is carried out on log data generated in real time during normal operation of the ATP system through the trained abnormal sensing model of the ATP system.
2. The method for real-time abnormality sensing of an automatic high-speed train protection system according to claim 1, wherein the calculation of the internal state using the forgetting gate, the input gate and the candidate state, and the calculation of the hidden state using the internal state and the output gate is expressed as: ; ; Wherein, the The internal state at time i is indicated, A forgetting gate indicating the moment i, The input gate at time i is indicated, An output gate at the moment i is indicated, Indicating a candidate state at time i, The product of the vector elements is represented, The internal state at time i-1 is indicated, Indicating the hidden state at time i.
3. The method for sensing real-time abnormality of automatic high-speed train protection system according to claim 1, wherein the hidden state of each moment of the decoding network, the hidden states of the last moment and the hidden states of all moments of the coding network are combined through the multi-head attention mechanism layer, and the calculated attention function determination comprises the following steps: Using a multi-head attention mechanism, each sub-device corresponds to an attention mechanism, and for the s-th sub-device, in the moment l, utilizing the hidden state corresponding to the moment l-1 As a query vector, the attention mechanism layer corresponding to the s-th sub-device is used for hiding the state of the coding network at all moments Calculating the attention function of the moment I corresponding to the s-th sub-device, and splicing the attention functions of the moment I corresponding to all the sub-devices to be used as the attention function of the moment I of the decoding network Reusing the attention function at time l Hidden state at time l-1 with decoding network Calculating hidden state of decoding network moment Wherein the hidden state of each moment of the decoding network comprises the hidden states corresponding to all the sub-devices.
4. A method for real-time anomaly perception of an automatic protection system for a high-speed train according to claim 3, wherein calculating an attention function by combining the hidden state of the last moment with the hidden states of all moments of the coding network by a multi-head attention mechanism layer comprises: For time l, attention function corresponding to s-th sub-device Calculated by the following formula: ; Wherein att (-) represents the attention mechanism layer, Representing hidden status at all moments, will Expressed as key value pairs , And To use parameter matrix in the corresponding attention mechanism of the s-th sub-device And (3) with The calculated key vector matrix and value vector matrix are calculated by the following steps: , , , , Is a key vector matrix Is used to determine the i-th key vector of (c), Is a vector matrix of values Is used to determine the value vector of the (i) th value vector, Represents the hidden state at the moment of the encoding network i, Represents the hidden state corresponding to the s-th sub-device l-1 moment, Representing the attention distribution corresponding to the s-th sub-equipment, wherein the input log data is a log sequence formed by log data generated by a plurality of sub-equipment in the ATP system, and m is the length of the input log sequence and is equal to the total time; The calculation formula of (2) is as follows: ; Wherein, the Representing an attention scoring function, calculated using a scaled dot product model, Is a key vector matrix Is the j-th key vector of (a); splicing the attention functions corresponding to all the sub-devices to obtain the attention function of the decoding network at the moment I 。
5. The real-time abnormality sensing method of the automatic high-speed train protection system according to any one of claims 1 to 4, wherein the ATP system abnormality sensing model comprises a log key abnormality sensing module and a parameter value abnormality sensing module, wherein the log key abnormality sensing module and the parameter value abnormality sensing module have the same structure and comprise a corresponding coding network, a decoding network, an attention mechanism layer and a classifier; Analyzing the log at the historical moment in advance to obtain two parts of log data, wherein one part is a log key and the other part is a vector parameter; the method comprises the steps of inputting log data consisting of log keys to a log key abnormality sensing module, training the log key abnormality sensing module in a training phase mode, inputting log data consisting of parameter value vectors to the parameter value abnormality sensing module, and training the parameter value abnormality sensing module in the training phase mode; And in the sensing stage, the log generated in real time during normal operation of the ATP system is analyzed, log data which is obtained by analysis and consists of log keys and log data which consists of vector parameters are respectively and correspondingly input into a trained log key abnormality sensing module and a trained parameter value abnormality sensing module, and when the output result of the log key abnormality sensing module or the parameter value abnormality sensing module is abnormal, the ATP system is determined to be abnormal.
6. The real-time abnormality sensing method of an automatic high-speed train protection system according to claim 5, wherein the abnormal log key sensing module is characterized in that in a training stage, km g epsilon K is used as a conditional probability distribution of a target log key, the abnormal log key sensing module is trained according to differences between the real log key and the real log key at corresponding time to update internal parameters of the abnormal log key sensing module, and the aim is to learn a conditional probability distribution capable of maximizing a training log key sequence, wherein K= { K 1 ,k 2 ,…,k u } represents a group of specified log keys of an ATP system, u is the number of the log keys, the first q log keys in the conditional probability distribution are marked as normal, and q is a positive integer for balancing abnormal sensing rate and false alarm rate; In the perception stage, for a target log key km g , based on an input log key window w, a conditional probability distribution Pt [ km g |w]={k 1 :p 1 , k 2 :p 2 ,…, k u :p u ] is calculated, wherein w= { km g-h ,…, km g-2 ,km g-1 } comprises h nearest log keys before the target log key km g , any element in w belongs to a set K, K 1 :p 1 , k 2 :p 2 ,…, k u :p u respectively represents that the probability corresponding to the log key K 1 is p 1 , the probability corresponding to the log key K 2 is p 2 , the probability corresponding to the log key K u is p u , the log keys are arranged in descending order according to the probability size, the first q log keys are determined to be corresponding candidate log keys, and if km g is within the q candidate log keys, the target log key km g is marked as normal, otherwise, the target log key km g is marked as abnormal.
7. The method for real-time anomaly sensing of an automatic high-speed train protection system according to claim 5, wherein the parameter value anomaly sensing module outputs a real-value vector as a predicted value of a next input vector parameter in a training phase, and reduces an error between the next input vector parameter and the predicted value of the next input vector parameter by dynamically adjusting weights of the LSTM model; in the sensing stage, in each moment, modeling the error between the input vector parameter and the predicted value as Gaussian distribution, if the error is within a confidence interval of the predicted value, the input vector parameter is marked as normal, otherwise, the input vector parameter is abnormal.
8. The method for real-time abnormality sensing of an automatic high-speed train protection system according to claim 5, wherein when the ATP system is determined to be abnormal in the sensing stage, a warning prompt is sent by expanding the display of the DMI to guide the user to execute the abnormality processing flow.
9. The real-time abnormality sensing method of the automatic high-speed train protection system according to claim 8, wherein different abnormality processing flows are abstracted into finite state automata, the system abnormality processing flows are depicted by state transition and execution sequences of the finite state automata, a knowledge base is formed on the basis, and the knowledge base is reused for the treatment of the same type of abnormality by all different operation and maintenance subjects.

Description

Real-time abnormality sensing method for automatic protection system of high-speed train Technical Field The invention relates to the technical field of rail transit, in particular to a real-time abnormality sensing method of an automatic protection system of a high-speed train. Background The train automatic protection system (Automatic Train Protection, ATP) is a core for ensuring safe and efficient operation of a high-speed train, and is called as a neural center of the high-speed train. The system is arranged at two ends of a high-Speed train, adopts a redundant structure, is connected with external equipment such as the train and a monitoring system, and mainly comprises a safety computer (Vital Computer, VC), a Speed and distance measuring Unit (Speed & Distance Processing Unit, SDU), a transponder information receiving Unit (Balise Transmission Module, BTM), a track circuit reader (Track Circuit Reader, TCR), a train interface (TRAIN INTERFACE Unit, TIU), a human-computer interface (Driver-MACHINE INTERFACE, DMI), a wireless transmission Unit (GSM-Railway, GSM-R), a judicial data recording Unit (Juridical Recorder Unit, JRU) and other sub-equipment. Each piece of sub equipment is matched with each other, so that the safety of the train is guaranteed. Failure of any of the sub-devices can affect the normal operation of the ATP system. By the end of 2022, there are 5 types of ATP systems for full-scale use, including 300T, 300S, 300H, 200H and 200C. After the system is put on line, each type of sub-equipment shows a plurality of fault modes, for example, the number of main accessories of a 300T type ATP system is 107, and single BTM related faults are divided into 5 types of runtime BSA (Balise Service Available, availability of transponder transmission service), startup BSA, invalid BTM ports, failure of correctly analyzing messages by BTMs, routine test failure and the like. It follows that there are thousands of possible failure modes for each type of ATP system, and that accurate fault localization is relatively difficult once the system fails. From the existing operation and maintenance status, the fault diagnosis and positioning of various ATP systems basically depend on manual work, and the problems of time consumption, low accuracy, low automation degree and the like of fault positioning exist. In addition, when an ATP system for executing a transport task breaks down on a line, the ATP system is usually temporarily handled by a driver or a vehicle-mounted mechanic, and there are cases in which a fault handling process is unfamiliar, a handling method is inappropriate, and a complex scene cannot be used, due to a barrier of professional skills, a blind area, and the like, the resulting fault delay and the scope of influence are large. According to different study objects and diagnosis modes, the fault diagnosis of ATP is mainly three methods, namely an empirical knowledge-based method, an analytical model-based method and a data driving-based method. With the development of new artificial intelligence technologies such as machine learning and the like and the requirement of fault diagnosis algorithms on the adaptability of ATP application scenes, a data-driven method becomes trend and mainstream. In recent years, scholars have proposed methods such as bayesian network, labeled-LDA (LATENT DIRICHLET Allocation), convolutional neural network, and extreme gradient lifting (extreme Gradient Boosting, XGBoost) for fault diagnosis and classification of ATP, although these methods have a certain effectiveness from the experimental results. However, from the view of the ATP practical application scenario, there is a timing correlation between the front and back operation data. In addition, the operation data are simultaneously and independently generated by a plurality of devices, and the fault location needs to be combined with information comprehensive judgment in multiple aspects in parallel, so that the concurrency is realized. Unfortunately, the above approach does not meet both the timing correlation and concurrency requirements of ATP fault diagnosis. Disclosure of Invention The invention aims to provide a real-time abnormality sensing method for an automatic protection system of a high-speed train, which can simultaneously meet the requirements of time sequence correlation, rapidness and concurrency of ATP fault diagnosis and accurately realize the real-time abnormality sensing of the automatic protection system of the high-speed train. The invention aims at realizing the following technical scheme: A real-time abnormality sensing method of an automatic protection system of a high-speed train comprises the following steps: Constructing an abnormal perception model of an ATP system, which comprises a coding network, a decoding network, an attention mechanism layer and a classifier, wherein the ATP system is an automatic protection system of a high-speed train; The method comprises the steps of p