Train forward operation risk assessment method based on operation risk domain

Abstract

The invention provides a train forward operation risk assessment method based on an operation risk domain, and the method proposes the concept of the operation risk domain based on a field theory, constructs a train forward operation risk assessment model based on the risk domain, and constructs operation risk domain models in two scenes of train tracking and foreign matter invasion. And the forward operation risk of the train is evaluated in real time, and management and control measures are given, so that the real-time and quantitative evaluation of the forward operation of the train is realized, and the construction of an autonomous safety guarantee system of rail transit is promoted.

Description

Technical Field

[0001] The present invention relates to the field of rail transit safety, and in particular to a train forward operation risk assessment method based on an operation risk domain. Background Art

[0002] Train operations are often affected by external environmental factors, particularly foreign object intrusion. These incidents are highly accidental, sudden, and difficult to predict, potentially leading to serious accidents such as derailments and collisions. Forward train operation risk refers to risks that may arise in the direction of travel, primarily from foreign objects or preceding trains. To ensure safe train operation and enhance the safety of railway system operations, assessing and managing forward train operation risk is crucial.

[0003] In the railway sector, safety risk assessment primarily relies on static risk analysis, which typically relies on historical data, experience, and expert knowledge. Static risk analysis involves identifying potential risk sources and assessing their potential consequences. Existing technologies employ the HAZOP method, combined with the ALARP risk matrix, to conduct qualitative risk analysis of subway signaling systems. Fault Tree Analysis (FTA), a semi-quantitative risk analysis method that introduces a tree structure to describe the interaction of multiple factors leading to accidents, has been used in railway transportation system risk assessment since the 1980s. The Analytic Hierarchy Process (AHP), proposed in the early 1970s, decomposes decision-making elements into multiple hierarchies, such as goals, criteria, and options, and conducts qualitative and quantitative analysis based on these factors. This study combines FTA and AHP to study railway safety risks, and describes the application of this method in detail using the high-speed train braking system as an example. Traditional risk assessment methods still rely primarily on empirical qualitative judgment and lack quantitative and real-time assessment methods.

[0004] In the field of autonomous driving, new risk analysis methods based on field models have garnered widespread attention. The Safety Force Field (SFF) model has been proposed. The SFF defines the trajectory left by an autonomous vehicle in space and time after adopting a safety strategy as a set of spatiotemporal declarations. By determining the overlap of spatiotemporal declarations of various road targets, the SFF model analyzes the collision risk of the autonomous vehicle, providing a theoretical basis for decision-making in various driving scenarios. Furthermore, based on field theory, a risk domain model is proposed to characterize driving risk using fields. This model, based on the concept of field theory, characterizes the impact of various human-vehicle-road factors on driving risk. By calculating the comprehensive "forces" applied to the vehicle within the risk domain, the SFF model is used to assess the driving risk at each location during the driving process. This model provides a quantitative and visual model capable of calculating driving risk in real time and predicting its dynamic changes, providing a theoretical basis for driving safety assistance algorithms and the implementation of safety decisions. The application of risk assessment and decision-making assistance methods based on field models in the field of autonomous driving has begun to receive widespread attention. At the same time, based on the driving risk domain model, by analyzing the interaction between risk domains, a driver risk response model under human-vehicle-road interaction is established. Road following experiments are also conducted to construct an assessment mechanism for the driver's status in human-machine co-driving and a model for allocating human-machine driving control rights. On this basis, an improved driving safety field model is developed, which combines the characteristics of autonomous perception, control, and decision-making of the unmanned vehicle AI system and improves the existing driving safety field model based on Hooke's law. Summary of the Invention

[0005] The train operation process is greatly affected by the external environment, among which the phenomenon of foreign body intrusion during forward operation often occurs, seriously affecting the safety of train operation. In order to solve the problems of qualitative, lagging, and subjective nature of the current operation risk assessment methods of the railway system, combined with the rapidly developing sensors and video monitoring systems, drones and other intelligent perception technologies, the present invention proposes a train forward operation risk assessment method based on the driving safety domain based on the field theory model. Realize the real-time, quantitative and visual risk assessment, and improve the utilization rate of measurable data in the field of risk assessment during train operation. The present invention specifically adopts the following technical solutions:

[0006] A method for evaluating train forward operation risk based on an operation risk domain comprises the following steps:

[0007] (1) Constructing a risk assessment model for forward train operation based on risk domain

[0008] (1.1) Calculation of operational risk domain intensity in train tracking scenarios

[0009] The risk source is divided into multiple risk elements according to their length, and the risk domain intensity generated by each risk element is calculated respectively. Finally, the integration calculation is performed to obtain the operational risk domain intensity generated by the preceding train:

[0010]

[0011]

[0012]

[0013] in: Indicates the intensity of the operational risk domain generated by the preceding train; Indicates the distance between the current running train and the preceding train; represents the risk weight of the forward train; is the length of the preceding train; is the comprehensive influence coefficient, which represents the influence of the train operation status and personnel, equipment and environmental factors during the train operation; is an undetermined coefficient, which represents the basic risk of the train and ensures the relative speed of the current train tracking operation hour ; 、 、 They are personnel, equipment, and environmental risk influencing factors; Indicates the relative speed between the current train and the preceding train; 、 、 Respectively represent the impact of three types of risk factors on the train during operation: personnel, equipment, and other external environmental factors;

[0014] (1.2) Calculation of operational risk domain intensity under foreign body intrusion scenarios

[0015] The risk sources in the foreign object intrusion scenario include foreign objects and personnel invading the line. Based on the risk size distribution calculation, the operational risk domain intensity in the foreign object intrusion scenario is obtained:

[0016]

[0017]

[0018] in:

[0019] Indicates the intensity of the risk domain caused by foreign matter invading the line; Indicates the risk weight of foreign objects or people invading the line; Indicates the distance between the current running train and the risk source; Indicates the relative speed between the current running train and the risk source; is the comprehensive impact coefficient;

[0020] (1.3) Calculation of train operation risk potential

[0021] The train operation risk potential is a measure of the risk of the train in the risk domain and is calculated as follows:

[0022]

[0023] in:

[0024] is the risk potential energy of the current running train; Indicates the distance between the current running train and the risk source; is the risk weight of the currently running train; The intensity of the risk zone where the current train is located;

[0025] Risk potential is directly used to assess the risk level of the currently running train, and the current train is directly classified into risk levels according to the size of the train's risk potential, and then control measures are formulated based on the risk assessment results.

[0026] Preferably, the method further comprises the steps of:

[0027] (2) Determine the model's undetermined parameters

[0028] (2.1) Determining risk weights

[0029] Risk sources are divided into three levels: I, II, and III, based on the severity of the train collision accidents they may cause. A moving train is classified as a Level I risk source, and in the case of foreign object intrusion, foreign objects are further categorized as Level II and Level III risk sources. A risk matrix approach is used for evaluation, classifying and scoring the consequences of different degrees to obtain risk weight scores for different degrees of severity.

[0030] (2.2) Determine risk influencing factors

[0031] The risk impact factors include personnel risk impact factors, equipment risk impact factors and environmental risk impact factors.

[0032] Preferably, the method further comprises the steps of:

[0033] (3) Solving the undetermined coefficient k

[0034] The risk levels are divided into low risk, general risk, high risk and major risk. The boundaries between two adjacent risk levels are defined. The tracking interval time and tracking distance of trains on a certain line are defined as the criteria for risk level classification.

[0035] set up Design speed for trains; is the minimum tracking interval of trains; The maximum braking distance of the train; the minimum tracking interval of the train As the basis for dividing low risk and general risk, the critical distance is calculated as ;Use the maximum braking distance of the train As the critical distance between greater risk and major risk; select and The average value of is used to divide general risk and greater risk, and the risk domain intensity calculation formula of step (1) is combined to obtain the unknown coefficient k.

[0036] Preferably, the method further comprises the steps of:

[0037] (4) Determine the risk level of the train

[0038] Undetermined coefficient After solving the problem, the critical values ​​of the risk potential energy corresponding to the three risk level boundaries of low risk and general risk, general risk and greater risk, and greater risk and major risk are calculated respectively. When conducting the forward train operation risk assessment, the corresponding risk domain model is selected according to the scenario, the risk potential energy of the EMU train under the current operating state is calculated, and compared with the critical value, the current risk level of the train can be directly obtained;

[0039] When the train is in a serious risk, the safety measure to be taken is a braking strategy to avoid accidents or minimize the losses caused by accidents;

[0040] When the train is at a high risk, it will reduce speed and apply brakes if necessary to prevent collisions with foreign objects that have not yet been cleared.

[0041] When the train status is at general risk, the dispatching department will promptly notify the work area to remove foreign objects and slow down the train if necessary. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a schematic diagram of the operation risk domain calculation in the train tracking scenario.

[0043] Figure 2 This is a schematic diagram of the calculation of the operational risk domain intensity under the foreign body intrusion scenario.

[0044] Figure 3 It is the risk potential diagram of the risk domain model.

[0045] Figure 4 It is a risk potential contour map.

[0046] Figure 5 It is a risk assessment flow chart. DETAILED DESCRIPTION

[0047] 1. Construction of the risk assessment model for forward train operation

[0048] 1.1 Definition of Operational Risk Domain

[0049] The operational risk domain refers to the distribution of the potential for different types and degrees of damage to a train caused by risk sources ahead of it during its operation. The risk domain model simultaneously describes the probability of accidents and the combination of damage, as well as their distribution along the track. The risk domain intensity at a point on the track represents the magnitude of the risk posed by the risk source at that point; the risk potential energy is the potential energy possessed by the train when it reaches a point in the risk domain. The risk potential energy represents the magnitude of the risk to the train at the current moment.

[0050] During train operation, the train is affected by multiple risk factors such as personnel, equipment, environment, and management system. Among them, the forward operation risk of the train is mainly affected by the environment. It is a risk event in the forward direction of the train operation, such as train collision and foreign object intrusion. Therefore, the two scenarios of train tracking and foreign object intrusion are used to construct the operation risk domain model respectively.

[0051] 1.2 Driving risk domain intensity

[0052] 1.2.1 Calculation of operational risk domain intensity in train tracking scenarios

[0053] In the train tracking scenario, the risk source is the preceding train. Since the train itself has a certain length, for example, the length of the CRH380BL EMU is 399.27m, the train itself cannot be regarded as a point mass. The risk distribution law generated by the preceding train is compared with the electric field distribution law generated by the charged straight line. Therefore, the idea of ​​infinitesimal elements is also used when calculating the operational risk domain intensity of a certain point on the line, such as Figure 1 As shown in the figure, the risk source is divided into risk elements according to their length, and the risk domain intensity generated by each risk element is calculated respectively. Finally, the integral calculation is performed to obtain the operational risk domain intensity formula generated by the preceding train:

[0054]

[0055]

[0056]

[0057] in:

[0058] Indicates the intensity of the operational risk domain generated by the preceding train;

[0059] Indicates the risk weight of the preceding train, i.e., the severity of the consequences of an accident between the train and the risk source;

[0060] is the length of the preceding train;

[0061] is the comprehensive influence coefficient, which represents the influence of the train operation status and personnel, equipment and environmental factors during the train operation;

[0062] is an undetermined coefficient, which represents the basic risk of the train, that is, to ensure the relative speed of the current train tracking operation hour ;

[0063] 、 、 They are personnel, equipment and environmental risk influencing factors respectively.

[0064] Indicates the relative speed between the current running train and the preceding train.

[0065] 、 、 It respectively represents the influence of three types of risk factors on the train during operation, namely personnel, equipment and other external environment. The value ranges of the three influencing factors are , the value is 1 when there is no impact, and the greater the impact on train operation, the larger the value.

[0066] 1.1.1 1.2.2 Calculation of operational risk domain strength in foreign body intrusion scenarios

[0067] In the foreign object intrusion scenario, the risk sources include foreign objects and people that invade the line. They can be regarded as particles relative to the entire railway line. Therefore, the risk distribution law is established according to the risk size, such as Figure 2 As shown:

[0068]

[0069]

[0070] in:

[0071] Indicates the intensity of the risk domain caused by foreign matter invading the line;

[0072] Indicates the risk weight of foreign objects or people invading the line;

[0073] Indicates the distance between the current running train and the risk source;

[0074] is the comprehensive impact coefficient;

[0075] 、 、 They are personnel, equipment and environmental risk influencing factors respectively.

[0076] Indicates the relative speed between the current running train and the risk source.

[0077] 1.3 Risk Potential

[0078] Train operation risk potential energy is a measure of the risk level of a train in the risk domain. Similar to the calculation method of electric field and electric potential energy, the current train risk level is related to the risk domain intensity, the distance between the risk source and the train, and the current train risk weight. The calculation formula is as follows:

[0079]

[0080] in:

[0081] is the risk potential energy of the current running train;

[0082] is the risk weight of the currently running train;

[0083] is the distance between the current running train and the risk source;

[0084] is the intensity of the risk zone where the current train is located, is a function of r.

[0085] Risk potential can be directly used to assess the risk level of the currently running train, and the current train can be directly classified into risk levels according to the size of the train's risk potential, and then control measures can be formulated based on the risk assessment results.

[0086] 2. Parameter determination and risk level classification

[0087] 2.1 Undetermined parameter values

[0088] 2.1.1 Risk Weights

[0089] Because the model scenarios involve train tracking and foreign object intrusion, the model primarily describes the risk of train collisions, including rear-end collisions, pedestrian collisions, collisions with foreign objects, and derailments. Therefore, in these two scenarios, the risk sources are the oncoming train and the intruding foreign object, respectively. Based on the severity of the train collision accident that a risk source could cause, the risk sources are classified into three levels: I, II, and III, as shown in Table 1.

[0090] Table 1 Types of risk sources

[0091]

[0092] Because rear-end collisions can lead to serious consequences such as mass casualties and rail traffic paralysis, the forward train is classified as a Level I risk source. In the foreign object intrusion scenario, foreign objects are further categorized into Level II and Level III risk sources. Level II risk sources include people illegally intruding on the track, falling rocks, and hard floating objects, which can create risks such as trains striking people or colliding with foreign objects, leading to train derailment. Level III risk sources include small animals and light objects that intrude on the railway clearances. These collisions can cause foreign objects to become lodged in the wheelset, damaging the train's running gear, bogies, and other equipment. There is also a certain risk of train derailment, and forcing the train to stop can affect the normal operation of the train group.

[0093] To characterize the risk weights of the three risk sources within the risk domain model, they need to be scored based on the severity of the incidents they cause. As shown in Table 2, the Railway Bureau categorizes and scores the consequences of different degrees when using the risk matrix method for evaluation, obtaining weights for different severity levels. Since risk sources are also graded based on the severity of the consequences, the risk weights of the three risk sources can be scored.

[0094] By comprehensively comparing the classification basis of Table 1 and Table 2, the risk weights of level I, II and III risk sources are respectively determined as follows: .

[0095] Table 2 Severity of hazardous incidents

[0096]

[0097] 2.1.2 Risk Influencing Factors

[0098] (1) Personnel risk influencing factors

[0099] The personnel risks during train operation mainly come from the human attributes. Different types of work have clear requirements for the physical and psychological qualities of personnel. Therefore, personnel risks can be divided into five components: safety literacy, ideological quality, professional quality, physiological quality and psychological quality. The evaluation scores of the five components are respectively Indicates that the personnel risk impact factor can be obtained:

[0100]

[0101] in, The value range of , it can be scored through real-time monitoring of people's working status, personnel physical examinations, assessments, etc. Under normal conditions, the value is 1. When the personnel have insufficient professional knowledge, physiological and psychological defects, etc., the value is greater than 1. The larger the value, the greater the impact of personnel risks on train operations.

[0102] (2) Equipment risk influencing factors

[0103] The equipment factors in the train operation process are complex and diverse, so it is necessary to classify the equipment and evaluate and score the equipment operation status according to the equipment category when calculating the equipment risk impact factor. The equipment in the train operation process can be divided into infrastructure, electrical facilities, communication equipment, EMU equipment, etc., and the equipment is classified into The evaluation scores of the four types of equipment can be used to calculate the equipment risk impact factors:

[0104]

[0105] in, The value range of The operating status of equipment in the railway system can be evaluated and scored by assessing the health status of the equipment and predicting the remaining service life of the equipment. If the equipment is operating normally, the evaluation score is 1. A larger score indicates a more serious decline in the health of the equipment and a greater impact on train operation.

[0106] (3) Environmental risk influencing factors

[0107] Environmental risk impact factors are used to assess external environmental impacts other than foreign object intrusion, and can therefore be divided into categories such as weather factors, disaster factors, and social environment. Weather factors such as rain and snow may lead to insufficient wheel-rail friction, thereby affecting the train's braking ability; disaster factors such as earthquakes and landslides may directly damage railway line equipment and interrupt train operations; social environmental factors such as garbage dumps and greenhouses around the railway will increase the intrusion of light objects such as films and entangle the contact network. Represent the assessment scores of weather, disasters, and social environment respectively, and calculate the environmental risk impact factor:

[0108]

[0109] in, The value range is During train operation, the high-speed railway's disaster monitoring system can monitor wind speed, precipitation, and earthquake conditions. In the future, the social environment around the railway can be monitored through drone inspections or video surveillance. The higher the evaluation score, the greater the impact of the environment on train operation.

[0110] In summary, the selection of influencing factors and the evaluation method of each score value still need further research. Therefore, the subsequent calculation and result output of the model assume that the train operation is not affected by abnormal personnel, equipment, and environmental factors, that is, With the development of sensing technologies such as sensors, video monitoring, and drones, more and more relevant measurable data will be available for calculating the risk factors affecting personnel, equipment, and the environment.

[0111] 2.1.3 Solving the Undetermined Coefficients

[0112] In the process of building the train operation risk domain model, there are still undetermined coefficients The solution has not yet been obtained. Risk potential is a numerical value used to characterize and assess the current operational risk level of a train. According to the railway department's definition and classification of risk, risk levels are divided into low risk, general risk, high risk, and major risk. Therefore, it is necessary to define the boundary between each two adjacent risk levels. The tracking interval and tracking distance of trains on a certain line are defined as the criteria for risk level classification, as shown in Table 3.

[0113] Table 3 Definition of risk level boundaries in the case of train tracking

[0114]

[0115] In the table:

[0116] Indicates the train's design speed;

[0117] Indicates the minimum tracking interval of trains;

[0118] Indicates the maximum braking distance of the train;

[0119] As shown in the table, the minimum tracking interval of trains is adopted As the basis for dividing low risk and general risk, the critical distance is calculated as ;Use the maximum braking distance of the train As the critical distance between greater risk and major risk; select and The average value of the risk factor is used to divide the risk into general and greater risks. Since the minimum tracking interval, maximum braking distance and design speed of the train are determined by the train model and line type, when conducting train operation risk assessment, it is necessary to define the risk level boundary for each type of train on the line to ensure that the train operation risk domain model corresponds to the train model.

[0120] Taking the CRH380BL model running on the Beijing-Shanghai High-Speed ​​Railway as an example, the design speed is 350km / h and the maximum braking distance is 6.5km. The minimum tracking interval of trains on the Beijing-Shanghai High-Speed ​​Railway during peak hours can be as short as 3 minutes. Therefore, according to Table 3, , , , so the critical distance of risk level is calculated as shown in Table 4. Through research and experience, it is found that in the actual operation of the railway system, hard floating objects that invade the track line generally need to be cleared within 60 minutes after being discovered, otherwise it will cause at least 60 minutes of delay, so it will be classified as a general Class D accident. The present invention takes the distance of 60 minutes of train running at a speed of 350km / h, that is, 350km, as the boundary distance between low risk and general risk in the risk domain model of Class II risk source. Therefore, the risk domain formula is combined to calculate the unknown coefficient .

[0121] Table 4 Risk level critical value table

[0122]

[0123] Undetermined coefficient Once these values ​​are known, the critical values ​​of the risk potential energy corresponding to the three risk level boundaries (low risk vs. general risk, general risk vs. high risk, and high risk vs. major risk) are calculated. When conducting forward train operation risk assessment, we simply select the corresponding risk domain model based on the scenario, calculate the risk potential energy of the EMU train's current operating state, and compare it with the critical value. This directly determines the risk level of the current train. Therefore, the risk domain model establishes a link between the train's operating state and risk assessment. Table 5 shows the values ​​of all parameters and variable value ranges for this model.

[0124] Table 5 Parameters and variable values ​​of train operation risk domain

[0125]

[0126] 2.2 Model results and risk level classification

[0127] According to the sensitivity analysis results, in the train operation risk domain model, the relative speed between the train and the risk source is and the distance between the train and the risk source The two variables contribute the most to the model and have the highest sensitivity. 、 is changing with time, so 、 As independent variables, the changing patterns of train operation risk potential energy with these two parameters are discussed and analyzed.

[0128] Figure 3 In 、 X-axis and Y-axis, with risk potential For the Z-axis, a 3D surface is drawn for a total of four risk domain assessment models for the two scenarios of train tracking and foreign object intrusion. The three-dimensional surface is used to characterize the relationship between the size of the risk potential energy and the change in relative speed and distance during the operation of the train. The four figures are the same in the distribution law and shape growth trend of the risk potential energy: when the distance from the risk source is the same, the higher the current train speed, the greater the risk potential energy; when the train speed is the same, the closer the distance between the train and the risk source, the greater the train operation risk potential energy. The main difference lies in the different levels of the risk source and the different relative positions to the line, which leads to different sizes and growth rates when calculating the risk potential energy: when 、 When the risk weight or risk impact factor of the risk source is the same, the greater the risk potential and the greater the growth rate. The risk level classification range of train operation is shown in Table 6. 、 According to the corresponding relationship with the risk level, the three sub-graphs are transformed into the contour map of risk potential, such as Figure 4 shown.

[0129] Table 6 Risk level classification and control measures

[0130]

[0131] like Figure 4 As shown in the figure, by setting three contour lines with values ​​of 40.0, 58.0, and 105.7, the map is divided into four areas in a clockwise direction: red, yellow, green, and blue, representing the ranges of major risk, large risk, general risk, and low risk respectively. The two-tuple coordinates represent the train running status, The color of the coordinate area represents the risk level of the train. Therefore, the risk potential contour map can intuitively show the risk level of the train in different scenarios and operating states, and take corresponding risk control measures according to Table 6.

[0132] When the train status is in the red zone, the train is at great risk and the possibility and severity of an accident are higher. The safety measures taken at this time are braking strategies to avoid accidents or minimize the losses caused by accidents.

[0133] When the train status is in the orange zone, the train is at great risk and the possibility and severity of an accident are high. Therefore, a speed reduction strategy is adopted and braking is performed when necessary to prevent collision with foreign objects that have not yet been cleared.

[0134] When a train is in the yellow zone, it is at general risk, indicating the possibility of an accident ahead. Therefore, the dispatching department needs to promptly notify the work area to remove foreign objects and, if necessary, reduce the train's speed. As the train slows down, its relative speed to the risk source also decreases, inhibiting the train's risk potential from moving to the upper left and promoting movement toward a lower risk potential. This reduces the risk level of the train and allows more time to clear intruding foreign objects. Therefore, the process of implementing risk control measures is a process of reversing the risk potential gradient.

[0135] 3. Model Application

[0136] The risk assessment application process based on the train operation risk domain is as follows: Figure 5 As shown in the figure, the basis and premise of the model operation is input data. With the development and application of sensing technologies such as sensors, video monitoring, lidar, and drones in railways, the main input of the model is relative speed. , distance from the risk source The measurability of the model has been greatly increased, and with the continuous advancement of technologies such as high-speed rail disaster prevention systems, human health monitoring, and equipment health management, the human risk influencing factors in the model have been greatly improved. , Equipment risk influencing factors , Environmental risk influencing factors The calculation method provides data and technical support, enabling the risk domain model to move beyond analyzing train collision risk and also assess the risks posed by factors such as personnel, equipment, and the environment during normal train tracking operations. For example, inclement weather like strong winds and heavy rain can be monitored in real time by the high-speed rail disaster prevention system's rainfall and wind speed monitoring equipment, updating the natural weather risk score. This can increase the environmental risk impact factor and the potential risk energy of train operation, leading to a higher risk level for the train's operating status.

[0137] Therefore, the above monitoring data is input and calculated through the train operation risk domain model to obtain the risk potential of train operation, evaluate the current train operation risk, calculate the risk level under the current operating state, and formulate corresponding risk control measures based on the risk level and risk factor composition.

[0138] After implementing risk control measures, the effectiveness of these measures requires a reassessment of the changed train operation risk. By inputting the post-control monitoring data, the system determines whether the risk level is low. If the risk level is low, the control measures are effective, and the risk assessment ceases and continues with the closed-loop iteration. If the risk level remains high, further measures are implemented, followed by a reassessment until the train operation risk is reduced to low, at which point the iteration cycle ceases.

[0139] The construction of a train operation risk domain model enables the application of monitoring data generated during train operation in train operation risk assessments, improving the efficiency of monitoring data utilization. Furthermore, the model can incorporate monitoring data on other risk factors, such as personnel, equipment, and the environment, expanding beyond the analysis of train collision risks and promoting the development of an integrated safety assurance system encompassing "sensing-recognition-judgment-action." The train operation risk domain can be used to assess the operational risk of a train at a given moment, thus being continuous in time. The magnitude of the train operation risk can be directly updated based on real-time monitoring data, making the analysis dynamic and real-time. Furthermore, based on the real-time risk assessment results, the temporal trend of operational risk can be mapped, facilitating risk prediction during train operation.

[0140] Aiming at the forward operation scenario of trains, the present invention comprehensively considers factors such as personnel, environment, and equipment, combines objective data with subjective experience, proposes the concept of operation risk domain, constructs a risk assessment model based on the operation risk domain, and performs model output analysis and risk level classification; constructs an integrated risk assessment process, making the train operation risk assessment method quantitative, real-time, visual, and integrated, providing new theories and research directions for the construction of an integrated "sensing-recognition-judgment-action" safety assurance system.

Claims

1. A train forward operation risk assessment method based on an operation risk domain, characterized in that: The method comprises the following steps: (1) Constructing a risk assessment model for forward train operation based on risk domain (1.1) Calculation of operational risk domain intensity in train tracking scenarios The risk source is divided into multiple risk elements according to their length, and the risk domain intensity generated by each risk element is calculated respectively. Finally, the integration calculation is performed to obtain the operational risk domain intensity generated by the preceding train: in: Indicates the intensity of the operational risk domain generated by the preceding train; Indicates the distance between the current running train and the preceding train; represents the risk weight of the forward train; is the length of the preceding train; is the comprehensive influence coefficient, which represents the influence of the train operation status and personnel, equipment and environmental factors during the train operation; is an undetermined coefficient, which represents the basic risk of the train and ensures the relative speed of the current train tracking operation hour ; 、 、 They are personnel, equipment, and environmental risk influencing factors; Indicates the relative speed between the current train and the preceding train; 、 、 Respectively represent the impact of three types of risk factors on the train during operation: personnel, equipment, and other external environmental factors; (1.2) Calculation of operational risk domain intensity under foreign body intrusion scenarios The risk sources in the foreign object intrusion scenario include foreign objects and personnel invading the line. Based on the risk size distribution calculation, the operational risk domain intensity in the foreign object intrusion scenario is obtained: in: Indicates the intensity of the risk domain caused by foreign matter invading the line; Indicates the risk weight of foreign objects or people invading the line; Indicates the distance between the current running train and the risk source; Indicates the relative speed between the current running train and the risk source; is the comprehensive impact coefficient; (1.3) Calculation of train operation risk potential The train operation risk potential is a measure of the risk of the train in the risk domain and is calculated as follows: in: is the risk potential energy of the current running train; Indicates the distance between the current running train and the risk source; is the risk weight of the currently running train; The intensity of the risk zone where the current train is located; Risk potential is directly used to assess the risk level of the currently running train, and the current train is directly classified into risk levels according to the size of the train's risk potential, and then control measures are formulated based on the risk assessment results.

2. The method for evaluating train forward operation risk based on an operation risk domain according to claim 1, wherein: The method further comprises the steps of: (2) Determine the model's undetermined parameters (2.1) Determining risk weights Risk sources are divided into three levels: I, II, and III, based on the severity of the train collision accidents they may cause. A moving train is classified as a Level I risk source, and in the case of foreign object intrusion, foreign objects are further categorized into Level II and Level III risk sources. The risk matrix method is used for evaluation, which classifies and scores the consequences of different degrees to obtain risk weight scores at different levels of severity; (2.2) Determine risk influencing factors The risk impact factors include personnel risk impact factors, equipment risk impact factors and environmental risk impact factors.

3. The method for evaluating train forward operation risk based on an operation risk domain according to claim 2, wherein: The method further comprises the steps of: (3) Solving the undetermined coefficient k The risk levels are divided into low risk, general risk, high risk and major risk. The boundaries between two adjacent risk levels are defined. The tracking interval time and tracking distance of trains on a certain line are defined as the criteria for risk level classification. set up Design speed for trains; is the minimum tracking interval of trains; The maximum braking distance of the train; the minimum tracking interval of the train As the basis for dividing low risk and general risk, the critical distance is calculated as ;Use the maximum braking distance of the train As the critical distance between greater risk and major risk; select and The average value of is used to divide general risk and greater risk, and the risk domain intensity calculation formula of step (1) is combined to obtain the unknown coefficient k.

4. The method for evaluating train forward operation risk based on an operation risk domain according to claim 3, characterized in that: The method further comprises the steps of: (4) Determine the risk level of the train Undetermined coefficient After solving the problem, the critical values ​​of the risk potential energy corresponding to the three risk level boundaries of low risk and general risk, general risk and greater risk, and greater risk and major risk are calculated respectively. When conducting the forward train operation risk assessment, the corresponding risk domain model is selected according to the scenario, the risk potential energy of the EMU train under the current operating state is calculated, and compared with the critical value, the current risk level of the train can be directly obtained; When the train is in a serious risk, the safety measure to be taken is a braking strategy to avoid accidents or minimize the losses caused by accidents; When the train is at a high risk, it will reduce speed and apply brakes if necessary to prevent collisions with foreign objects that have not yet been cleared. When the train status is at general risk, the dispatching department will promptly notify the work area to remove foreign objects and slow down the train if necessary.

Images