An intelligent urban rail transit emergency risk assessment method and system

Abstract

The present invention discloses an intelligent urban rail transit emergency risk assessment method and system, the method including risk factor mapping, construction of an emergency risk assessment model and urban rail transit emergency risk assessment. The present invention belongs to the field of data processing, and specifically refers to an intelligent urban rail transit emergency risk assessment method and system. This scheme introduces a cosine smoothing design activation function to accurately capture subtle changes in risk indicators in a local area; it can make the transition of risk changes softer and more stable under a stable state; through detailed regional division and integral averaging, it integrates local risk information into the overall risk function, thereby more accurately reflecting the local risk state in the overall assessment, helping to timely locate emergency risk events; reducing the noise in sensor data and the impact of collection interruptions; and thus improving the final risk assessment effect.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent urban rail transit emergency risk assessment method and system. Background Art

[0002] The urban rail transit emergency risk assessment method is a comprehensive approach that utilizes historical and real-time data, maps risk factors, and quantitatively analyzes various risk factors to achieve dynamic assessment and early warning of emergency risks. However, typical urban rail transit emergency risk assessment methods are unable to adapt to the rapid changes in risk indicators during emergencies and are insufficiently sensitive to slowly changing risk indicators in urban rail transit, resulting in low accuracy in the final risk assessment. They are also insufficiently sensitive to measurement errors and discontinuous data collection, ignoring local risk differences across different areas and stations within urban rail transit, making it difficult to promptly reflect local emergencies, which in turn leads to poor results in the final risk assessment. Summary of the Invention

[0003] In view of the above situation, in order to overcome the defects of the existing technology, the present invention provides an intelligent urban rail transit emergency risk assessment method and system. In view of the problem that the general urban rail transit emergency risk assessment method cannot adapt to the rapid changes of risk indicators in emergencies and is not sensitive enough to the slowly changing risk indicators in urban rail transit, which leads to low accuracy of the final risk assessment, this solution introduces cosine smoothing to design an activation function, which accurately captures subtle changes in risk indicators in local areas, so that when anomalies occur in local areas of stations and lines, it can respond in time; it can make the transition of risk changes softer and more stable in a stable state, thereby improving the accuracy of urban rail transit system. The solution improves the ability to handle slowly changing risk indicators in the system and improves the accuracy of risk assessment. In view of the fact that general urban rail transit emergency risk assessment methods are insufficiently sensitive to measurement errors and discontinuous data collection, ignore the local risk differences in different areas and stations in urban rail transit, and are unable to reflect local emergencies in a timely manner, which in turn leads to poor final risk assessment results, this solution integrates local risk information into the overall risk function through detailed regional division and integral averaging, thereby more accurately reflecting the local risk status in the overall assessment, helping to locate sudden risk events in a timely manner, reducing the impact of noise and collection interruptions in sensor data, and thus improving the final risk assessment effect.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides an intelligent urban rail transit emergency risk assessment method, which includes the following steps:

[0005] Step S1: risk factor mapping;

[0006] Step S2: Constructing an emergency risk assessment model;

[0007] Step S3: Urban rail transit emergency risk assessment.

[0008] Furthermore, in step S1, the risk factor mapping is to obtain historical data and risk assessment levels related to various emergencies in urban rail transit; use the risk assessment levels as data labels; and normalize each risk factor and map it to a region, expressed as: ;in, is the total sampling area obtained by normalizing all risk factors from all collected data; s is the total number of risk factors, is the total number of collected data dimensions; r is the dimension index; and They are the lower and upper bounds of the normalized risk factor; the risk factor is the dimension of the collected data.

[0009] Furthermore, in step S2, the construction of the emergency risk assessment model is to embed a local integral smoothing layer on the basis of the neural network; specifically, the following steps are included:

[0010] Step S21: Basic network structure design; design a multi-layer feedforward network, the input layer accepts the feature vector x, and outputs the risk assessment score after passing through N layers of hidden layers. The network parameters are recorded as , the initial risk assessment function for ;

[0011] Step S22: Activation function design; construct multivariate activation function , expressed as: ; ;in, is the activation function input; s is the total number of dimensions, and r is the dimension index; is the basic activation function; is the activation function input in the rth dimension; a and z are width parameters; is the activation contribution weight; and β control the parameters of the slope and midpoint, respectively;

[0012] Step S23: local integral smoothing layer design; based on the preliminary network output, the local integral mean is introduced to average the risk values in each small area; and the final emergency risk assessment model output is obtained; including:

[0013] Step S231: Region division; normalized region The risk factors are evenly divided into Sub-regions, each sub-region represents a continuous area in the data space, rather than the value of a single data point, and includes all the data that fall within the area; the coordinate axis of each sub-region corresponds to a risk factor; each sub-region Defined as: ;in, is the discretized index in the rth dimension; n+1 is the interval divided into n+1 parts in each dimension;

[0014] Step S232: Local integral mean construction; for each sub-region The risk function value within is integrated and averaged, which is expressed as: ;in, is the integral function, u is the integral variable; and are the lower bounds of the 1st and sth dimensions respectively; and are the upper bounds of the 1st and sth dimensions respectively; is the step-by-step integration variable;

[0015] Step S233: Use the local integral mean to locally approximate the overall risk function. The final risk assessment output is defined as: ;in, is the scale factor; h is the normalized area The step size of the upper segmentation; are grid points uniformly distributed within the sub-area; It is the output of risk assessment for the sampling point; is the index of the sub-area divided on the first risk factor dimension; is the index of the sub-region divided on the s-th risk factor dimension.

[0016] Furthermore, in step S3, the urban rail transit emergency risk assessment is to perform risk assessment on the urban rail transit emergency data collected in real time based on the established emergency risk assessment model.

[0017] The present invention provides an intelligent urban rail transit emergency risk assessment system, which includes a risk factor mapping module, an emergency risk assessment model construction module and an urban rail transit emergency risk assessment module;

[0018] The risk factor mapping module obtains historical data and risk assessment levels related to various emergencies in urban rail transit; and constructs a risk factor mapping;

[0019] The emergency risk assessment model construction module embeds a local integral smoothing layer on the basis of a neural network, and realizes the construction of an emergency risk assessment model by designing a multivariate activation function and a local integral mean technology;

[0020] The urban rail transit emergency event risk assessment module performs risk assessment on urban rail transit emergency event data collected in real time based on the constructed risk assessment model.

[0021] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0022] (1) In view of the fact that general urban rail transit emergency risk assessment methods are unable to adapt to the rapid changes of risk indicators in emergencies and are not sensitive enough to slowly changing risk indicators in urban rail transit, which leads to low accuracy of the final risk assessment, this scheme introduces cosine smoothing to design an activation function, which accurately captures subtle changes in risk indicators in local areas, so that when anomalies occur in local areas of stations and lines, timely responses can be made; it can make the transition of risk changes softer and more stable under stable conditions, thereby improving the ability to handle slowly changing risk indicators in urban rail transit systems and improving the accuracy of risk assessment.

[0023] (2) In view of the fact that general urban rail transit emergency risk assessment methods are not sensitive enough to measurement errors and discontinuous data collection, and ignore the local risk differences in different areas and stations in urban rail transit, it is difficult to reflect local emergencies in a timely manner, which leads to poor final risk assessment results. This scheme integrates local risk information into the overall risk function through detailed regional division and integral averaging, thereby more accurately reflecting the local risk status in the overall assessment, helping to locate emergency risk events in a timely manner; reducing the noise in sensor data and the impact of data collection interruptions; and thus improving the final risk assessment effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flow chart of a method for risk assessment of intelligent urban rail transit emergencies provided by the present invention;

[0025] Figure 2 A schematic diagram of an intelligent urban rail transit emergency risk assessment system provided by the present invention;

[0026] Figure 3 Schematic diagram of the process of step S2.

[0027] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0029] In the description of the present invention, it should be understood that terms such as "up", "down", "front", "back", "left", "right", "top", "bottom", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the system or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they should not be understood as limiting the present invention.

[0030] Example 1, see Figure 1 The present invention provides an intelligent urban rail transit emergency risk assessment method, which includes the following steps:

[0031] Step S1: Risk factor mapping; obtaining historical data and risk assessment levels related to various types of emergencies in urban rail transit; constructing a risk factor mapping;

[0032] Step S2: Constructing an emergency risk assessment model; on the basis of the neural network, embedding a local integral smoothing layer, and realizing the construction of the emergency risk assessment model by designing a multivariate activation function and a local integral mean technology;

[0033] Step S3: Risk assessment of urban rail transit emergencies; risk assessment is performed based on the constructed risk assessment model.

[0034] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, risk factor mapping is to obtain historical data and risk assessment levels related to various types of emergencies in urban rail transit; the risk assessment levels are used as data labels; the historical data related to various types of emergencies in urban rail transit include the status of the rail signal system, train running times, service intervals, delays, crowd density, passenger flow, equipment status, and emergency response time; the risk assessment levels include mild danger, moderate danger, and severe danger; each risk factor is normalized and mapped to a region, expressed as: ;in, is the total sampling area obtained by normalizing all risk factors from all collected data; s is the total number of risk factors, is the total number of collected data dimensions; r is the dimension index; and They are the lower and upper bounds of the normalized risk factor; the risk factor is the dimension of the collected data.

[0035] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, the emergency risk assessment model is constructed by embedding a local integral smoothing layer on the basis of a neural network to achieve local data smoothing, noise elimination, and local approximation of the risk function. Specifically, the following steps are included:

[0036] Step S21: Basic network structure design: Design a basic neural network, the input layer accepts the feature vector x, and outputs the risk assessment score after passing through N hidden layers. The network parameters are recorded as , the initial risk assessment function for ;

[0037] Step S22: Activation function design; construct multivariate activation function , expressed as: ; ;in, is the activation function input; s is the total number of dimensions, and r is the dimension index; is the basic activation function; is the activation function input in the rth dimension; a and z are width parameters; is the activation contribution weight; and β control the parameters of slope and midpoint, respectively;

[0038] Step S23: local integral smoothing layer design.

[0039] By performing the above operations, we can address the problem that general urban rail transit emergency risk assessment methods are unable to adapt to the rapid changes in risk indicators in emergencies and are insufficiently sensitive to slowly changing risk indicators in urban rail transit, which leads to low accuracy in the final risk assessment. This solution introduces cosine smoothing to design an activation function, which accurately captures subtle changes in risk indicators in local areas, allowing for timely responses when anomalies occur in local areas of stations and lines; it can make the transition of risk changes softer and more stable under stable conditions, thereby improving the ability to handle slowly changing risk indicators in the urban rail transit system and improving the accuracy of risk assessment.

[0040] Example 4, see Figure 1 This embodiment is based on the above embodiment. In step S23, the local integral smoothing layer is designed to deal with the measurement errors and data collection discontinuity problems existing in urban rail transit. Based on the preliminary network output, the local integral mean is introduced to average the risk values in each small area; the final emergency risk assessment model output is obtained; including:

[0041] Step S231: Region division; normalized region The risk factors are evenly divided into Sub-regions, each sub-region represents a continuous area in the data space, rather than the value of a single data point, and includes all the data that fall within the area; the coordinate axis of each sub-region corresponds to a risk factor; each sub-region Defined as: ;in, is the discretized index in the rth dimension; n+1 is the interval divided into n+1 parts in each dimension;

[0042] Step S232: Local integral mean construction; for each sub-region The risk function value within is integrated and averaged, which is expressed as: ;in, is the integral function, u is the integral variable; and are the lower bounds of the 1st and sth dimensions respectively; and are the upper bounds of the 1st and sth dimensions respectively; is the step-by-step integration variable;

[0043] Step S233: Use the local integral mean to locally approximate the overall risk function. The final risk assessment output is defined as: ;in, is the scale factor; h is the normalized area The step size of the upper segmentation; are grid points uniformly distributed within the sub-area; It is the output of risk assessment for the sampling point; is the index of the sub-area divided on the first risk factor dimension; is the index of the sub-region divided on the s-th risk factor dimension.

[0044] By performing the above operations, we can address the problem that general urban rail transit emergency risk assessment methods are insufficiently sensitive to measurement errors and discontinuous data collection, ignore the local risk differences in different areas and stations in urban rail transit, and have difficulty in timely reflecting local emergencies, which in turn leads to poor final risk assessment results. This solution integrates local risk information into the overall risk function through detailed regional division and integral averaging, thereby more accurately reflecting the local risk status in the overall assessment, helping to timely locate emergency risk events; reduce the impact of noise and collection interruptions in sensor data; and thus improve the final risk assessment effect.

[0045] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S3, the urban rail transit emergency risk assessment is based on the established emergency risk assessment model to conduct risk assessment on the real-time collected urban rail transit emergency data. If the model output is a serious danger, an early warning is issued to the relevant personnel.

[0046] Example 6, see Figure 2 , this embodiment is based on the above embodiment, and the present invention provides an intelligent urban rail transit emergency risk assessment system, including a risk factor mapping module, an emergency risk assessment model construction module and an urban rail transit emergency risk assessment module;

[0047] The risk factor mapping module obtains historical data and risk assessment levels related to various emergencies in urban rail transit; and constructs a risk factor mapping;

[0048] The emergency risk assessment model construction module embeds a local integral smoothing layer on the basis of a neural network, and realizes the construction of an emergency risk assessment model by designing a multivariate activation function and a local integral mean technology;

[0049] The urban rail transit emergency event risk assessment module performs risk assessment on urban rail transit emergency event data collected in real time based on the constructed risk assessment model.

[0050] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0051] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions, and alterations can be made to the embodiments without departing from the principles and spirit of the invention.

[0052] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. A method for risk assessment of emergencies in intelligent urban rail transit, characterized by: The method comprises the following steps: Step S1: Risk factor mapping; obtaining historical data and risk assessment levels related to various emergencies in urban rail transit; constructing a risk factor mapping; Step S2: Constructing an emergency risk assessment model; on the basis of the neural network, embedding a local integral smoothing layer, and realizing the construction of the emergency risk assessment model by designing a multivariate activation function and a local integral mean technology; Step S3: Urban rail transit emergency risk assessment: risk assessment is performed based on the constructed risk assessment model; In step S1, the risk factor mapping is to obtain historical data and risk assessment levels related to various emergencies in urban rail transit; use the risk assessment levels as data labels; and normalize each risk factor and map it to a region, expressed as: ;in, is the total sampling area obtained by normalizing all risk factors from all collected data; s is the total number of risk factors, is the total number of collected data dimensions; r is the dimension index; and are the lower and upper bounds of the normalized risk factor, respectively; the risk factor is the dimension of the collected data; In step S2, the construction of the emergency risk assessment model is to embed a local integral smoothing layer on the basis of the neural network; specifically, the following steps are included: Step S21: Basic network structure design; design a multi-layer feedforward network, the input layer accepts the feature vector x, and outputs the risk assessment score after passing through N layers of hidden layers. The network parameters are recorded as , the initial risk assessment function for ; Step S22: Activation function design; construct multivariate activation function , expressed as: ; ;in, is the activation function input; s is the total number of dimensions, and r is the dimension index; is the basic activation function; is the activation function input in the rth dimension; a and z are width parameters; is the activation contribution weight; and β control the parameters of the slope and midpoint, respectively; Step S23: local integral smoothing layer design; based on the preliminary network output, the local integral mean is introduced to average the risk values in each small area; and the final emergency risk assessment model output is obtained; including: Step S231: Region division; normalized region The risk factors are evenly divided into Sub-regions, each sub-region represents a continuous area in the data space, rather than the value of a single data point, and includes all the data that fall within the area; the coordinate axis of each sub-region corresponds to a risk factor; each sub-region Defined as: ;in, is the discretized index in the rth dimension; n+1 is the interval divided into n+1 parts in each dimension; Step S232: Local integral mean construction; for each sub-region The risk function value within is integrated and averaged, which is expressed as: ;in, is the integral function, u is the integral variable; and are the lower bounds of the 1st and sth dimensions respectively; and are the upper bounds of the 1st and sth dimensions respectively; is the step-by-step integration variable; Step S233: Use the local integral mean to locally approximate the overall risk function. The final risk assessment output is defined as: ;in, is the scale factor; h is the normalized area The step size of the upper segmentation; are grid points uniformly distributed within the sub-area; It is the output of risk assessment for the sampling point; is the index of the sub-area divided on the first risk factor dimension; is the index of the sub-region divided on the s-th risk factor dimension.

2. The intelligent urban rail transit emergency risk assessment method according to claim 1, characterized in that: In step S3, the urban rail transit emergency risk assessment is to perform risk assessment on the urban rail transit emergency data collected in real time based on the established emergency risk assessment model.

3. An intelligent urban rail transit emergency risk assessment system, used to implement an intelligent urban rail transit emergency risk assessment method according to any one of claims 1-2, characterized in that: It includes risk factor mapping module, emergency risk assessment model building module and urban rail transit emergency risk assessment module; The risk factor mapping module obtains historical data and risk assessment levels related to various types of emergencies in urban rail transit; construct risk factor mapping; The emergency risk assessment model construction module embeds a local integral smoothing layer on the basis of a neural network, and realizes the construction of an emergency risk assessment model by designing a multivariate activation function and a local integral mean technology; The urban rail transit emergency event risk assessment module performs risk assessment on urban rail transit emergency event data collected in real time based on the constructed risk assessment model.

Images