CN115187048A - Method and system for identifying the situation of foreign body intrusion event in multi-domain boundary of track line - Google Patents

Abstract

The invention provides a method and a system for identifying the condition of a foreign object intrusion event in a multi-domain boundary of a track line, belonging to the technical field of railway disaster prevention risk early warning. Collect; divide the orbital boundary area into attention domain, early warning domain, and safety domain, identify, track, and extract features of foreign objects that invade the boundary; define the state description model of the intrusion event; The characteristic data of intrusion-limited foreign objects are used as the dynamic characteristic parameters of the intrusion event state description model to quantify the risk value, and the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain is calculated. The invention realizes the monitoring, identification and analysis of track line intrusion event characteristics and event occurrence process, further improves the risk prevention and control capability of the track transit system, and has good application and popularization value.

Description

Method and system for identification of foreign body intrusion event status in multi-domain boundary of track line

technical field

The invention relates to the technical field of railway disaster prevention risk early warning, and in particular relates to a method and a system for recognizing the condition of a foreign object intrusion event in a multi-domain boundary of a track line.

Background technique

The topography and landforms along the railway line vary greatly, and the geological conditions are complex. The foreign body intrusion event may lead to serious consequences such as line failure, train delay, and even train derailment.

The boundary area of the rail surface is a contour line delimited to limit the insurmountability of foreign objects in order to protect the safety of rail line transportation. At present, in view of the problem of foreign body intrusion in the track line, the existing technology mainly adopts the mode of "recognition + alarm" by using dual-cable sensors, microwave monitoring sensors, fiber grating sensors or monitoring video in the boundary area of the track surface. , that is, to detect whether there is a foreign body intrusion limit, if a foreign body is found, an alarm will be triggered directly, and then manual intervention will be performed.

The existing technology has a direct detection method in practical applications, with fast detection speed but low alarm effectiveness, but also has the disadvantage of not being able to identify the characteristics of the intruded object, and at the same time, due to the imprecise division of the boundary area, it is difficult to analyze the dynamic change of the intrusion foreign object. It is necessary to rely on manual experience to judge the danger level of foreign body intrusion events and the development trend of intrusion events; on the other hand, the existing foreign body intrusion detection methods can only identify and alarm the first foreign body intrusion in the boundary area, and need manual processing and analysis. It can work again only after manual reset, and it cannot be recognized for the occurrence of secondary foreign body intrusion, resulting in low efficiency in the identification and processing of track line intrusion events.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a track line multi-domain boundary foreign object intrusion event that realizes the monitoring, identification and analysis of the characteristics of the multi-domain intrusion event and the event occurrence process of the rail line, and improves the risk prevention and control ability of the rail transit system. A situation identification method and system are provided to solve at least one technical problem existing in the above-mentioned background art.

In order to achieve the above object, the present invention has adopted the following technical solutions:

On the one hand, the present invention provides a method for identifying the condition of a multi-domain boundary foreign body intrusion event of a track line, including:

According to the monitoring requirements of the whole process of the occurrence of foreign body intrusion limit events, the track surface state acquisition sensor equipment is selected to collect the track surface state data. The collected data types and data formats are different. According to the characteristics of the original data collected, data analysis and data prediction processing, data enhancement operations;

Collect sensor data types according to different orbital states, analyze different data characteristics of the boundary area, divide the orbital boundary area into attention domain, early warning domain, and safety domain, and identify, track and feature extraction of foreign objects that invade the boundary;

Define an intrusion event state description model, take the identified intrusion event state description model as the dynamic feature of the intrusion event state description model, and take the track type and track condition as the static feature of the intrusion event state description model;

Based on the state characteristics of the intrusion event, the identified intrusion foreign object feature data is used as the dynamic characteristic parameter of the intrusion event state description model to quantify the risk value, and the intrusion events occurring in the concern domain, early warning domain and safety domain are proposed according to their risk degree. The level, early warning level, and risk early warning classification under the safety level are used to calculate the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain.

Preferably, when setting the track surface state acquisition sensor device, obtain the data under the safety state of the monitored track line and obtain the track surface boundary area in response, and divide the track surface boundary area into a safety domain, an early warning domain, and a concern domain, wherein The safety domain is the outline dimension line of the railway track, and the warning domain and the attention domain are expanded outwards; the neural network model is used to identify the foreign objects and their types; further, to identify and judge whether the foreign objects are in the safety domain, the warning domain or the attention domain, if yes Carry out frame selection on the foreign object target, and further determine whether the foreign object target has been tracked; otherwise, continue to obtain the track line status data to be identified;

If the foreign object has been tracked, calculate, update, and obtain the size data, moving speed data, moving direction, intrusion duration data, and limit position data of the intruding foreign object as the multi-level feature of the intruding foreign object; otherwise, Add the foreign object target into the tracker, and obtain the plane image of the track line to be identified again for identification;

It is judged that the data validity test is performed on the data of the foreign object type, the size data of the intrusion-limited foreign object, the moving speed data, the moving direction, the data of the intrusion limit time, and the limit position data.

Preferably, the encroachment event state description model is composed of static features and dynamic features, wherein the static features include track types and environmental risk types, and the dynamic features include foreign object attributes and foreign object movement behavior attributes. In the static features, the track type describes the characteristics of the track line itself, and the geological environment type describes the scene where the track line is located. and the size of the foreign body, where the size of the foreign body is the plane surface area, plane projection area, and volume in the multi-level features of the foreign body described above; the motion behavior of the foreign body describes the dynamic change of the intrusion event, which is specifically manifested as the intrusion of the intrusion of foreign objects into different orbits. The trend of the boundary area of the surface, such as the change of the moving speed of the foreign body, the movement direction of the foreign body, the change of the duration of the intrusion limit, and the change of the boundary position occupied.

Preferably, the real-time risk assessment model based on the state of the intrusion event includes:

First, determine the weights of the parameters of the state description model of the intrusion event, and face the different requirements of foreign object intrusion detection in the track line, and determine the track type weight W _RT , the geological environment type W _ET , and the foreign object attribute weight of the track line through expert experience. W _OA , foreign body motion attribute weight W _OMA ;

Analyze the track type risk quantification score RS _RT and the geological environment type risk quantification score RS _ET of the static features; According to the multi-level features of real-time intrusion and limiting foreign bodies, from the foreign body type and foreign body size data, the foreign body attribute risk quantification score RS _OAs is synthesized; by The foreign object moving speed, intrusion time duration, and the limit position data are used to synthesize the risk quantification score RS _OMA of foreign object movement attributes; the real-time risk value calculation model of the intrusion event is defined as shown in the following formula, and the static characteristics and dynamic characteristics of the model are described by inputting the intrusion event state. The weight of each indicator and the quantitative risk score of the feature are used to obtain the real-time risk value of the violation event;

R=W _RT ×RS _RT +W _ET ×RS _ET +W _0A ×RS _OA +W _OMA ×RS _OMA .

Preferably, in the method for calculating the detection rate of intrusion-limited foreign body events in the orbital boundary safety zone, early warning zone, and attention zone, the detection rate is focused on the occurrence of foreign body intrusion events within a period of time, indicating the foreign bodies in different boundary areas. How many intrusion events are detected correctly and early warning, the calculation formula is defined as follows:

The detection rate of the attention level is the ratio of the number of detected events to all the events that occurred in the intrusion events that occurred in the area of interest. The calculation method is as follows:

The detection rate of an early warning level is the ratio of the number of detected incidents to all incidents in the intrusion events that occurred in the early warning domain. The calculation method is as follows:

The detection rate of the security level is the ratio of the number of detected events to all the events that occurred in the security domain. The calculation method is as follows:

Preferably, the refined feature analysis of intruded foreign objects based on orbital multi-domain live data includes:

Step S201: Obtain the data under the safety state of the monitored track line and correspondingly, divide the track boundary area into a safety domain, an early warning domain, and a concern domain, and obtain the data characteristics of the track boundary area, wherein the safety domain is the rail outline dimension line, The early warning domain and concern domain expand outwards accordingly;

Step S202: input the track line state data after data preprocessing and data enhancement;

Step S203: Use models such as neural network and support vector machine to identify the type features of the invading foreign body, and perform frame selection;

Step S204: judging the area where the invading foreign object is located, and when the foreign object is in the area of interest, start tracking the foreign object target;

Step S205: Compare and match the type features and frame selection features of the intrusion-limited foreign object with the target tracker record to determine whether the foreign object has been tracked; if the foreign object has been tracked, go to step S207, otherwise go to step S207 S206;

Step S206: adding the foreign object target to the target tracker record, and returning to step S202 to obtain the next frame of data;

Step S207: Calculate and update the size data of the intrusion-limited foreign object;

Step S208: Calculate and update the movement direction data of the intrusion-limited foreign object;

Step S209: Calculate and update the foreign object moving speed data;

Step S210: Calculate and update the foreign body intrusion time limit data;

Step S211: Calculate and update the limit position data where the foreign object is located;

Step S212: judging to perform a data validity test on the foreign object type data, intrusion-limited foreign object size data, moving speed data, intrusion-limited duration data, and limit position data, and if the inspection passes, go to step S213, otherwise go to step S214;

Step S213: if the data is valid, output the multi-level feature of the invading foreign object;

Step S214: the data is invalid, and the frame data is discarded.

In a second aspect, the present invention provides a multi-domain boundary foreign object intrusion event status identification system for a track line, including:

The processing module is used to select the track surface state acquisition sensor equipment to collect the track surface state data according to the monitoring requirements of the whole process of the occurrence of the foreign body intrusion limit event. The collected data types and data formats are different. Data analysis, data preprocessing, and data enhancement operations;

The extraction module is used to collect sensor data types according to different rail surface states, analyze different data characteristics of the boundary area, divide the rail boundary area into attention domain, early warning domain, and safety domain, and identify, track, and characterize foreign objects that invade the boundary. extract;

The definition module is used to define the state description model of the intrusion event, and the identified intrusion foreign body features are regarded as the dynamic characteristics of the intrusion event state description model, and the track type and track condition are regarded as the static characteristics of the intrusion event state description model;

The calculation module is used to quantify the risk value based on the state characteristics of the intrusion event and use the identified characteristic data of the intrusion foreign object as the dynamic characteristic parameter of the state description model of the intrusion event. Its risk level proposes the concern level, early warning level, and risk early warning level under the safety level, and calculates the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain.

In a third aspect, the present invention provides a computer device comprising a memory and a processor, the processor and the memory are in communication with each other, the memory stores program instructions executable by the processor, the processor calls The program instructions execute the above-mentioned method for identifying the condition of a foreign object intrusion event in a multi-domain boundary of a track line.

In a fourth aspect, the present invention provides an electronic device, comprising a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, and the processor calls The program instructions execute the above-mentioned method for identifying the condition of a foreign object intrusion event in a multi-domain boundary of a track line.

In a fifth aspect, the present invention provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, implements the above-mentioned method for identifying a situation of a foreign object intrusion in a multi-domain boundary of a track line.

The invention has the beneficial effects that the monitoring, identification and analysis of the track line intrusion event characteristics and the event occurrence process are realized, and the risk prevention and control capability of the track transit system is further improved, which has good application and promotion value.

Additional aspects and advantages of the present invention will be set forth in part in the following description, which will be apparent from the following description, or may be learned by practice of the present invention.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

FIG. 1 is a flow chart of a method for identifying and analyzing a risk of a foreign object intrusion event in a multi-domain boundary of a track line according to an embodiment of the present invention.

FIG. 2 is a flow chart of a method for fine-grained feature analysis of intrusive foreign objects based on orbital multi-domain live data according to an embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below through the accompanying drawings are exemplary and are only used to explain the present invention, but not to be construed as a limitation of the present invention.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined as herein, are not to be taken in an idealized or overly formal sense. explain.

It will be understood by those skilled in the art that the singular forms "a", "an", "the" and "the" as used herein can include the plural forms as well, unless expressly stated otherwise. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of stated features, integers, steps, operations, elements and/or components, but does not exclude the presence or addition of one or more other features, Integers, steps, operations, elements and/or groups thereof.

In the description of this specification, description with reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples", etc., mean specific features described in connection with the embodiment or example , structure, material or feature is included in at least one embodiment or example of the present invention. Furthermore, the particular features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, those skilled in the art may combine and combine the different embodiments or examples described in this specification, as well as the features of the different embodiments or examples, without conflicting each other.

In order to facilitate the understanding of the present invention, the present invention will be further explained and described below with reference to the accompanying drawings with specific embodiments, and the specific embodiments do not constitute limitations to the embodiments of the present invention.

Those skilled in the art should understand that the accompanying drawings are only schematic diagrams of the embodiments, and the components in the accompanying drawings are not necessarily necessary to implement the present invention.

Example 1

The present embodiment 1 provides a multi-domain boundary foreign body intrusion event status identification system for a track line, and the system includes:

The processing module is used to select the track surface state acquisition sensor equipment to collect the track surface state data according to the monitoring requirements of the whole process of the occurrence of the foreign body intrusion limit event. The collected data types and data formats are different. Data analysis, data preprocessing, and data enhancement operations;

The extraction module is used to collect sensor data types according to different rail surface states, analyze different data characteristics of the boundary area, divide the rail boundary area into attention domain, early warning domain, and safety domain, and identify, track, and characterize foreign objects that invade the boundary. extract;

The definition module is used to define the state description model of the intrusion event, and the identified intrusion foreign body features are regarded as the dynamic characteristics of the intrusion event state description model, and the track type and track condition are regarded as the static characteristics of the intrusion event state description model;

The calculation module is used to quantify the risk value based on the state characteristics of the intrusion event and use the identified characteristic data of the intrusion foreign object as the dynamic characteristic parameter of the state description model of the intrusion event. Its risk level proposes the concern level, early warning level, and risk early warning level under the safety level, and calculates the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain.

In the present embodiment 1, the above-mentioned system is used to realize a method for identifying the condition of a multi-domain boundary foreign body intrusion event of a track line, including:

According to the monitoring requirements of the whole process of the occurrence of foreign body intrusion limit events, the track surface state acquisition sensor equipment is selected to collect the track surface state data. The collected data types and data formats are different. According to the characteristics of the original data collected, data analysis and data prediction processing, data enhancement operations;

Collect sensor data types according to different orbital states, analyze different data characteristics of the boundary area, divide the orbital boundary area into attention domain, early warning domain, and safety domain, and identify, track and feature extraction of foreign objects that invade the boundary;

Define an intrusion event state description model, take the identified intrusion event state description model as the dynamic feature of the intrusion event state description model, and take the track type and track condition as the static feature of the intrusion event state description model;

Based on the state characteristics of the intrusion event, the identified intrusion foreign object feature data is used as the dynamic characteristic parameter of the intrusion event state description model to quantify the risk value, and the intrusion events occurring in the concern domain, early warning domain and safety domain are proposed according to their risk degree. The level, early warning level, and risk early warning classification under the safety level are used to calculate the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain.

When setting up the track surface state acquisition sensor equipment, obtain the data under the safety state of the monitored track line and obtain its track surface boundary area in response, and divide the track surface boundary area into safety domain, early warning domain, and attention domain, of which the safety domain is The outline dimension line of the railway track, the warning domain and the attention domain are expanded outwards; the neural network model is used to identify the foreign objects and their types; further, identify and judge that the foreign objects are in the safety domain, the warning domain or the attention domain, and if so, the foreign bodies will be identified. The target is framed to further determine whether the foreign object has been tracked; otherwise, continue to obtain the track line status data to be identified;

If the foreign object has been tracked, calculate, update, and obtain the size data, moving speed data, moving direction, intrusion duration data, and limit position data of the intruding foreign object as the multi-level feature of the intruding foreign object; otherwise, Add the foreign object target into the tracker, and obtain the plane image of the track line to be identified again for identification;

It is judged that the data validity test is performed on the data of the foreign object type, the size data of the intrusion-limited foreign object, the moving speed data, the moving direction, the data of the intrusion limit time, and the limit position data.

The state description model of the encroachment event is composed of static features and dynamic features. The static features include track types and environmental risk types, and the dynamic features include foreign object attributes and foreign object movement behavior attributes. In the static features, the track type describes the characteristics of the track line itself, and the geological environment type describes the scene where the track line is located. and the size of the foreign body, where the size of the foreign body is the plane surface area, plane projection area, and volume in the multi-level features of the foreign body described above; the motion behavior of the foreign body describes the dynamic change of the intrusion event, which is specifically manifested as the intrusion of the intrusion of foreign objects into different orbits. The trend of the boundary area of the surface, such as the change of the moving speed of the foreign body, the movement direction of the foreign body, the change of the duration of the intrusion limit, and the change of the boundary position occupied.

A real-time risk assessment model based on the status of intrusion events, including:

First, determine the weights of the parameters of the state description model of the intrusion event, and face the different requirements of foreign object intrusion detection in the track line, and determine the track type weight W _RT , the geological environment type W _ET , and the foreign object attribute weight of the track line through expert experience. W _OA , foreign body motion attribute weight W _OMA ;

Analyze the track type risk quantification score RS _RT and the geological environment type risk quantification score RS _ET of the static features; According to the multi-level features of real-time intrusion and limiting foreign bodies, from the foreign body type and foreign body size data, the foreign body attribute risk quantification score RS _OAs is synthesized; by The foreign object moving speed, intrusion time duration, and the limit position data are used to synthesize the risk quantification score RS _OMA of foreign object movement attributes; the real-time risk value calculation model of the intrusion event is defined as shown in the following formula, and the static characteristics and dynamic characteristics of the model are described by inputting the intrusion event state. The weight of each indicator and the quantitative risk score of the feature are used to obtain the real-time risk value of the violation event;

R=W _RT ×RS _RT +W _ET ×RS _ET +W _0A ×RS _OA +W _OMA ×RS _OMA .

The method for calculating the detection rate of intrusion-limited foreign body events in the orbital boundary safety zone, early warning zone and attention zone, the detection rate is focused on the occurrence of foreign body intrusion events within a period of time, indicating the foreign body intrusion events in different boundary areas. How many are detected correctly and warned, the calculation formula is defined as follows:

The detection rate of the attention level is the ratio of the number of detected events to all the events that occurred in the intrusion events that occurred in the area of interest. The calculation method is as follows:

The detection rate of an early warning level is the ratio of the number of detected incidents to all incidents in the intrusion events that occurred in the early warning domain. The calculation method is as follows:

The detection rate of the security level is the ratio of the number of detected events to all the events that occurred in the security domain. The calculation method is as follows:

Refinement feature analysis of intrusive foreign objects based on multi-domain live data of orbits, including:

Step S201: Obtain the data under the safety state of the monitored track line and correspondingly, divide the track boundary area into a safety domain, an early warning domain, and a concern domain, and obtain the data characteristics of the track boundary area, wherein the safety domain is the rail outline dimension line, The early warning domain and concern domain expand outwards accordingly;

Step S202: input the track line state data after data preprocessing and data enhancement;

Step S203: Use models such as neural network and support vector machine to identify the type features of the invading foreign body, and perform frame selection;

Step S204: judging the area where the invading foreign object is located, and when the foreign object is in the area of interest, start tracking the foreign object target;

Step S205: Compare and match the type features and frame selection features of the intrusion-limited foreign object with the target tracker record to determine whether the foreign object has been tracked; if the foreign object has been tracked, go to step S207, otherwise go to step S207 S206;

Step S206: adding the foreign object target to the target tracker record, and returning to step S202 to obtain the next frame of data;

Step S207: Calculate and update the size data of the intrusion-limited foreign object;

Step S208: Calculate and update the movement direction data of the intrusion-limited foreign object;

Step S209: Calculate and update the foreign object moving speed data;

Step S210: Calculate and update the foreign body intrusion time limit data;

Step S211: Calculate and update the limit position data where the foreign object is located;

Step S212: judging to perform a data validity test on the foreign object type data, intrusion-limited foreign object size data, moving speed data, intrusion-limited duration data, and limit position data, and if the inspection passes, go to step S213, otherwise go to step S214;

Step S213: if the data is valid, output the multi-level feature of the invading foreign object;

Step S214: the data is invalid, and the frame data is discarded.

Example 2

Referring to FIG. 1 , the present embodiment 2 provides a method for identifying and risk analysis of a multi-domain boundary foreign body intrusion event of a track line, and the specific method includes:

Step S101: Set track surface state acquisition sensor equipment in the key monitoring area of the track line. In order to meet the whole-process monitoring requirements for the occurrence of foreign body intrusion events, the track surface state acquisition sensor equipment needs to have the ability to (1) distinguish the track boundary area from the sensing data , (2) the function of obtaining the plane or three-dimensional data of the intrusion and limiting foreign objects. In practical applications, any track surface state acquisition sensor device can be selected to operate independently or in combination according to the track line scene conditions and monitoring requirements.

Step S102: For different types of track surface state acquisition sensor devices, the data types and data formats collected are different, and data analysis, data preprocessing, and data enhancement operations are performed according to the original data characteristics collected by the track surface state acquisition sensor devices, and for image types The data is cropped, deblurred, and dehazed; segmentation, noise reduction, and downsampling are performed for point cloud type data. Further, the track line state data to be identified is obtained, and the process proceeds to step S103.

Step S103: When setting the track surface state acquisition sensor device, obtain the data under the safety state of the monitored track line and obtain its track surface boundary area in response, and divide the track surface boundary area into a safety domain, an early warning domain, and a concern domain, wherein The safety domain is the outline dimension line of the railway track, and the warning domain and the attention domain expand outward from this.

Next, input the track line status data to be identified in step S102, and use the neural network model to identify the foreign objects and their types; further, identify and judge whether the foreign objects are in the area of interest, and if so, select the foreign object target, and further Determine whether the foreign object has been tracked; otherwise, continue to acquire the track line state data to be identified in step S102.

If the foreign object has been tracked, further calculate, update, and obtain the size data (plane surface area, plane projection area, volume), moving speed data, moving direction data, intrusion time duration data, and limited area data of the intruding foreign object. As the multi-level feature of the intruded foreign object; otherwise, the foreign object target is added into the tracker, and the plane image of the track line to be identified in step S102 is acquired again for identification.

Further, it is judged that the data validity test is performed on the foreign body type data, the size data of the foreign body intrusion limit, the moving speed data, the time limit data of the invasion limit, and the data of the limit position. If the test is passed, the process goes to step S104, otherwise it is determined that the data is invalid and discarded. data.

Step S104: When a foreign object intrusion occurs on the track line, the risk of the intrusion event has a certain correlation with the intrusion of the foreign object, the environment in which the track is located, and the type of the track. In order to quantitatively evaluate the degree of intrusion risk, the state description model of the intrusion event is defined as follows. The state description model of the intrusion event is composed of static features and dynamic features. Foreign body movement behavior properties. In the static features, the track type describes the characteristics of the track line itself, and the geological environment type describes the scene where the track line is located. and the size of the foreign body, where the size of the foreign body is the plane surface area, plane projection area, and volume in the multi-level features of the foreign body described above; the motion behavior of the foreign body describes the dynamic change of the intrusion event, which is specifically manifested as the intrusion of the intrusion of foreign objects into different orbits. The trend of the boundary area of the surface, such as the change of the moving speed of the foreign body, the movement direction of the foreign body, the change of the intrusion time, and the change of the boundary position occupied. Among them, the moving direction of the foreign object describes the movement of the intrusion-limited foreign object to the safe area, or to the direction outside the area of interest. As shown in Table 1, the model parameters are described for the state of the encroachment event.

Table 1

Step S105: Construct a real-time risk assessment model based on the state of the intrusion event, first determine the weights of the parameters of the description model of the intrusion event state, and face different requirements for the detection of foreign body intrusion of the track line, and determine the risk of the track line through expert experience. Track type weight W _RT , geological environment type W _ET , foreign body attribute weight W _OA , foreign body motion attribute weight W _OMA .

Secondly, the track type risk quantification score RS _RT and the geological environment type risk quantification score RS _ET of the static feature are analyzed.

Further, according to the multi-level features of real-time intrusion and limiting foreign objects output in steps S103, S104, and S105, the foreign object attribute risk quantification score RS _OAs is synthesized from the foreign object type and foreign object size data; Boundary location data synthesized the risk quantification score RS _OMA for foreign body motion attributes. Proceed to step S106.

The possibility of intrusion-limited foreign objects leading to rail traffic accidents is different in different rail surface boundary areas. When the corresponding intrusion-limited foreign objects are located in different rail surface boundary areas, the intrusion-limited event state pays attention to different intrusion-limited foreign body characteristic parameters. When the foreign object is in the area of concern, the risk level belongs to the level of concern, and the foreign object movement behavior attribute focuses on evaluating the changes in the moving speed and moving direction of the foreign object. When entering the early warning domain, the risk level belongs to the early warning level, focusing on evaluating all the foreign body movement behavior attributes, and assessing whether the risk of the intrusion event has the possibility to be further improved; With reference to all foreign body motion behavior attributes, the impact of intrusion events on the safe operation of railways is evaluated.

Step S106 : define a limit violation event real-time risk value calculation model as shown in Equation 1, and obtain the limit violation event real-time risk value by inputting the static features of the limit violation event state description model, various index weights of dynamic features, and risk quantification scores.

R=W _RT ×RS _RT +W _ET ×RS _ET +W _OA xRS _OA +W _OMA ×RS _OMA

Table 2 The risk value of the limit violation event and its corresponding risk level division

Further, the real-time risk value of the intrusion event is normalized. According to Table 2, according to the real-time risk value of the intrusion event, the early warning interval is analyzed, and the track intrusion risk level is determined.

Step S107: Verify the detection rate of intrusion-limited foreign object events in the orbital-bounded safety domain, early warning domain, and attention domain. In this method, the detection rate is used to evaluate the accuracy of early warning of limit violation events. The detection rate focuses on the concentration of foreign body invasion events within a period of time, indicating how many foreign body invasion events in different boundary areas are correctly detected and warned. The detection rate of the attention level is the ratio of the number of detected events to all the events that occurred in the intrusion events that occurred in the area of interest. The calculation method is as follows:

The detection rate of an early warning level is the ratio of the number of detected incidents to all incidents in the intrusion events that occurred in the early warning domain. The calculation method is as follows:

The detection rate of the security level is the ratio of the number of detected events to all the events that occurred in the security domain. The calculation method is as follows:

Referring to Fig. 2, a method for analyzing the refined features of intrusive foreign objects based on orbital multi-domain live data, the specific steps include:

Step S201: Obtain the data under the safety state of the monitored track line and correspondingly, divide the track boundary area into a safety domain, an early warning domain, and a concern domain, and obtain the data characteristics of the track boundary area, wherein the safety domain is the rail outline dimension line, The early warning domain and the concern domain expand outwards accordingly. ;

Step S202: input the track line state data after data preprocessing and data enhancement;

Step S203: Use models such as neural network and support vector machine to identify the type features of the invading foreign body, and perform frame selection;

Step S204: judging the area where the invading foreign object is located, and when the foreign object is in the area of interest, start tracking the foreign object target;

Step S205: Compare and match the type feature and frame selection feature of the intrusion-limited foreign object with the record of the target tracker, and further, determine whether the foreign object target has been tracked; if the foreign object target has been tracked, proceed to step S207, otherwise will enter step S206;

Step S206: adding the foreign object target to the target tracker record, and returning to step S202 to obtain the next frame of data;

Step S207: Calculate and update the size data (plane surface area, plane projected area, volume) of the intrusion-limited foreign object;

Step S208: Calculate and update the movement direction data of the intrusion-limited foreign objects

Step S209: Calculate and update the foreign object moving speed data;

Step S210: Calculate and update the foreign body intrusion time limit data;

Step S211: Calculate and update the limit position data where the foreign object is located;

Step S212: Further, it is judged that the data validity test is performed on the foreign body type data, the size data of the foreign body intrusion limit, the moving speed data, the data of the time limit of the invasion limit, and the data of the limit position. If the test is passed, go to step S213, otherwise go to step S213 S214;

Step S213: if the data is valid, output the multi-level feature of the invading foreign object;

Step S214: the data is invalid, and the frame data is discarded.

Example 3

Embodiment 3 of the present invention provides an electronic device, including a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor invokes the The described program instruction executes the method for identifying the condition of the multi-domain boundary foreign object intrusion event of the track line, and the method includes the following process steps:

According to the monitoring requirements of the whole process of the occurrence of foreign body intrusion limit events, the track surface state acquisition sensor equipment is selected to collect the track surface state data. The collected data types and data formats are different. According to the characteristics of the original data collected, data analysis and data prediction processing, data enhancement operations;

Collect sensor data types according to different orbital states, analyze different data characteristics of the boundary area, divide the orbital boundary area into attention domain, early warning domain, and safety domain, and identify, track and feature extraction of foreign objects that invade the boundary;

Define an intrusion event state description model, take the identified intrusion event state description model as the dynamic feature of the intrusion event state description model, and take the track type and track condition as the static feature of the intrusion event state description model;

Based on the state characteristics of the intrusion event, the identified intrusion foreign object feature data is used as the dynamic characteristic parameter of the intrusion event state description model to quantify the risk value, and the intrusion events occurring in the concern domain, early warning domain and safety domain are proposed according to their risk degree. The level, early warning level, and risk early warning classification under the safety level are used to calculate the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain.

Example 4

Embodiment 4 of the present invention provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, realizes a method for identifying a condition of a multi-domain boundary foreign object intrusion event of a track line, and the method includes the following process steps:

According to the monitoring requirements of the whole process of the occurrence of foreign body intrusion limit events, the track surface state acquisition sensor equipment is selected to collect the track surface state data. The collected data types and data formats are different. According to the characteristics of the original data collected, data analysis and data prediction processing, data enhancement operations;

Collect sensor data types according to different orbital states, analyze different data characteristics of the boundary area, divide the orbital boundary area into attention domain, early warning domain, and safety domain, and identify, track and feature extraction of foreign objects that invade the boundary;

Define an intrusion event state description model, take the identified intrusion event state description model as the dynamic feature of the intrusion event state description model, and take the track type and track condition as the static feature of the intrusion event state description model;

Based on the state characteristics of the intrusion event, the identified intrusion foreign object feature data is used as the dynamic characteristic parameter of the intrusion event state description model to quantify the risk value, and the intrusion events occurring in the concern domain, early warning domain and safety domain are proposed according to their risk degree. The level, early warning level, and risk early warning classification under the safety level are used to calculate the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain.

Example 5

Embodiment 5 of the present invention provides a computer device, including a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor calls the The described program instruction executes the method for identifying the condition of the multi-domain boundary foreign body intrusion event of the track line, and the method includes the following steps:

According to the monitoring requirements of the whole process of the occurrence of foreign body intrusion limit events, the track surface state acquisition sensor equipment is selected to collect the track surface state data. The collected data types and data formats are different. According to the characteristics of the original data collected, data analysis and data prediction processing, data enhancement operations;

Collect sensor data types according to different orbital states, analyze different data characteristics of the boundary area, divide the orbital boundary area into attention domain, early warning domain, and safety domain, and identify, track and feature extraction of foreign objects that invade the boundary;

Define an intrusion event state description model, take the identified intrusion event state description model as the dynamic feature of the intrusion event state description model, and take the track type and track condition as the static feature of the intrusion event state description model;

Based on the state characteristics of the intrusion event, the identified intrusion foreign object feature data is used as the dynamic characteristic parameter of the intrusion event state description model to quantify the risk value, and the intrusion events occurring in the concern domain, early warning domain and safety domain are proposed according to their risk degree. The level, early warning level, and risk early warning classification under the safety level are used to calculate the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain.

To sum up, the embodiment of the present invention provides the method for identifying and analyzing the risk of foreign matter intrusion events in a multi-domain boundary of a track line. Aiming at the key problems such as the low processing efficiency of the intrusion event caused by the existing track foreign body intrusion monitoring method, the lack of identification of the intrusion foreign body characteristics and the effective monitoring of the intrusion event occurrence process, the status identification and risk of the track foreign body intrusion event are designed. The analysis method is to collect sensor properties and data characteristics for different orbital states, and propose a multi-level feature of intrusion event state and its identification method based on the attention domain, early warning domain and safety domain of the orbital boundary region. Risk assessment to achieve accurate risk early warning of intrusion events. Firstly, the real-time state data of the track surface is collected through the track surface state acquisition sensor device; secondly, data enhancement preprocessing is performed for the data types of the different track surface state acquisition sensors, and the type, size, intrusion limit time, moving speed, Refinement features of moving direction and intrusion location; based on the refined features of intrusion foreign objects and the requirements of track line safety, a state description model of intrusion events is constructed to describe the occurrence situation and key information of intrusion events; According to the event characteristics, a risk assessment model of the track foreign object intrusion event is constructed, the risk value of the intrusion event is obtained, and the risk early warning classification method under the concern level, early warning level and safety level is proposed according to the risk degree of the intrusion event. It realizes the monitoring, identification and analysis of the characteristics of the track line intrusion event and the event occurrence process, and further improves the risk prevention and control ability of the rail transit system, which has good application and promotion value.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing apparatus, where a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions for execution on the computer or other programmable apparatus Steps are provided for implementing the functions specified in a flow or flows of the flowcharts and/or a block or blocks of the block diagrams.

Although the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, they do not limit the scope of protection of the present invention. Those skilled in the art should understand that on the basis of the technical solutions disclosed in the present invention, those skilled in the art do not need to pay Various modifications or deformations that can be made by creative work shall be covered within the protection scope of the present invention.

Claims (10)

1. A method for identifying the situation of a multi-domain bounded foreign body intrusion event of a track line, is characterized in that, comprises: According to the monitoring requirements of the whole process of the occurrence of foreign body intrusion limit events, the track surface state acquisition sensor equipment is selected to collect the track surface state data. The collected data types and data formats are different. According to the characteristics of the original data collected, data analysis and data prediction processing, data enhancement operations; Collect sensor data types according to different orbital states, analyze different data characteristics of the boundary area, divide the orbital boundary area into attention domain, early warning domain, and safety domain, and identify, track and feature extraction of foreign objects that invade the boundary; Define an intrusion event state description model, take the identified intrusion event state description model as the dynamic feature of the intrusion event state description model, and take the track type and track condition as the static feature of the intrusion event state description model; Based on the state characteristics of the intrusion event, the identified intrusion foreign object feature data is used as the dynamic characteristic parameter of the intrusion event state description model to quantify the risk value, and the intrusion events occurring in the concern domain, early warning domain and safety domain are proposed according to their risk degree. The level, early warning level, and risk early warning classification under the safety level are used to calculate the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain. 2. The method for recognizing the condition of a multi-domain boundary foreign body intrusion event condition of a track line according to claim 1, wherein when the track surface state acquisition sensor device is set, the data under the monitored track line safety state and the acquisition of the response are obtained. The track boundary area is divided into safety domain, early warning domain, and attention domain. The safety domain is the outline dimension line of the rail, and the warning domain and attention domain expand outwards. The neural network model is used to identify foreign objects that appear. and its type; further, identify and judge that the foreign object is in the safety domain, early warning domain or concern domain, if yes, frame the foreign object target, and further judge whether the foreign object target has been tracked; otherwise, continue to obtain the track line to be identified status data; If the foreign object has been tracked, calculate, update, and obtain the size data, moving speed data, moving direction, intrusion duration data, and limit position data of the intruding foreign object as the multi-level feature of the intruding foreign object; otherwise, Add the foreign object target into the tracker, and obtain the plane image of the track line to be identified again for identification; It is judged that the data validity test is performed on the data of the foreign object type, the size data of the intrusion-limited foreign object, the moving speed data, the moving direction, the data of the intrusion limit time, and the limit position data. 3. The method for recognizing the state of a multi-domain boundary foreign body intrusion event of a track line according to claim 1, wherein the intrusion event state description model is composed of static features and dynamic features, wherein the static features include track types, environmental risk types , the dynamic features include foreign body attributes and foreign body movement behavior attributes. In the static features, the track type describes the characteristics of the track line itself, and the geological environment type describes the scene where the track line is located. and the size of the foreign body, where the size of the foreign body is the plane surface area, plane projection area, and volume in the multi-level features of the foreign body described above; the motion behavior of the foreign body describes the dynamic change of the intrusion event, which is specifically manifested as the intrusion of the intrusion of foreign objects into different orbits. The trend of the boundary area of the surface, such as the change of the moving speed of the foreign body, the movement direction of the foreign body, the change of the duration of the intrusion limit, and the change of the boundary position occupied. 4. The method for recognizing the state of an intrusion event of a multi-domain boundary foreign body in a track line according to claim 1, wherein the real-time risk assessment model based on the state of the intrusion event comprises: First, determine the weights of the parameters of the state description model of the intrusion event, and face the different requirements of foreign object intrusion detection in the track line, and determine the track type weight W _RT , the geological environment type W _ET , and the foreign object attribute weight of the track line through expert experience. W _OA , foreign body motion attribute weight W _OMA ; Analyze the track type risk quantification score RS _RT and the geological environment type risk quantification score RS _ET of the static features; According to the multi-level features of real-time intrusion and limiting foreign bodies, from the foreign body type and foreign body size data, the foreign body attribute risk quantification score RS _OAs is synthesized; by The foreign object moving speed, intrusion time duration, and the limit position data are used to synthesize the risk quantification score RS _OMA of foreign object movement attributes; the real-time risk value calculation model of the intrusion event is defined as shown in the following formula, and the static characteristics and dynamic characteristics of the model are described by inputting the intrusion event state. The weight of each indicator and the quantitative risk score of the feature are used to obtain the real-time risk value of the violation event; R=W _RT ×RS _RT +W _ET ×RS _ET +W _OA ×RS _OA +W _OMA ×RS _OMA . 5. The method for recognizing the state of foreign object intrusion events in a multi-domain boundary of a track line according to claim 1, characterized in that, the method for calculating the detection rate of intrusion-limited foreign body events in the described rail surface boundary safety domain, early warning domain, and attention domain , the detection rate is focused on the occurrence of foreign body intrusion events within a period of time, indicating how many foreign body intrusion events in different boundary areas are correctly detected and warned. The calculation formula is defined as follows: The detection rate of the attention level is the ratio of the number of detected events to all the events that occurred in the intrusion events that occurred in the area of interest. The calculation method is as follows:

The detection rate of an early warning level is the ratio of the number of detected incidents to all incidents in the intrusion events that occurred in the early warning domain. The calculation method is as follows:

The detection rate of the security level is the ratio of the number of detected events to all the events that occurred in the security domain. The calculation method is as follows:

6. The method for identifying the state of a foreign body intrusion event in a multi-domain boundary of a track line according to claim 1, wherein the refined feature analysis of the intrusion-limited foreign body based on the multi-domain live data of the track comprises: Step S201: Obtain the data under the safety state of the monitored track line and correspondingly, divide the track boundary area into a safety domain, an early warning domain, and a concern domain, and obtain the data characteristics of the track boundary area, wherein the safety domain is the rail outline dimension line, The early warning domain and concern domain expand outwards accordingly; Step S202: input the track line state data after data preprocessing and data enhancement; Step S203: Use models such as neural network and support vector machine to identify the type features of the invading foreign body, and perform frame selection; Step S204: judging the area where the invading foreign object is located, and when the foreign object is in the area of interest, start tracking the foreign object target; Step S205: Compare and match the type features and frame selection features of the intrusion-limited foreign object with the target tracker record to determine whether the foreign object has been tracked; if the foreign object has been tracked, go to step S207, otherwise go to step S207 S206; Step S206: adding the foreign object target to the target tracker record, and returning to step S202 to obtain the next frame of data; Step S207: Calculate and update the size data of the intrusion-limited foreign object; Step S208: Calculate and update the movement direction data of the intrusion-limited foreign object; Step S209: Calculate and update the foreign object moving speed data; Step S210: Calculate and update the foreign body intrusion time limit data; Step S211: Calculate and update the limit position data where the foreign object is located; Step S212: judging to perform a data validity test on the foreign object type data, intrusion-limited foreign object size data, moving speed data, intrusion-limited duration data, and limit position data, and if the inspection passes, go to step S213, otherwise go to step S214; Step S213: if the data is valid, output the multi-level feature of the invading foreign object; Step S214: the data is invalid, and the frame data is discarded. 7. A track line multi-domain boundary foreign body intrusion event status identification system, characterized in that, comprising: The processing module is used to select the track surface state acquisition sensor equipment to collect the track surface state data according to the monitoring requirements of the whole process of the occurrence of the foreign body intrusion limit event. The collected data types and data formats are different. Data analysis, data preprocessing, and data enhancement operations; The extraction module is used to collect sensor data types according to different rail surface states, analyze different data characteristics of the boundary area, divide the rail boundary area into attention domain, early warning domain, and safety domain, and identify, track, and characterize foreign objects that invade the boundary. extract; The definition module is used to define the state description model of the intrusion event, and the identified intrusion foreign body features are regarded as the dynamic characteristics of the intrusion event state description model, and the track type and track condition are regarded as the static characteristics of the intrusion event state description model; The calculation module is used to quantify the risk value based on the state characteristics of the intrusion event and use the identified characteristic data of the intrusion foreign object as the dynamic characteristic parameter of the state description model of the intrusion event. Its risk level proposes the concern level, early warning level, and risk early warning level under the safety level, and calculates the detection rate of intrusion-limited foreign object events in the orbital boundary safety domain, early warning domain, and attention domain. 8. A computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the multi-domain boundary foreign body intrusion limit of the track line according to any one of claims 1-6 is realized Incident condition identification method. 9. A computer device comprising a memory and a processor, the processor and the memory being in communication with each other, the memory storing program instructions executable by the processor, the processor invoking the program instructions to execute The method for identifying the condition of a foreign object intrusion event in a multi-domain boundary of a track line according to any one of claims 1 to 6. 10. An electronic device, comprising a memory and a processor, wherein the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, and the processor invokes the The program instruction executes the method for recognizing the condition of a foreign object intrusion event in a multi-domain boundary of a track line according to any one of claims 1-6.

Images