Method and system for constructing high-speed railway line risk map by considering transportation efficiency loss

Abstract

The invention relates to the technical field of traffic transportation engineering, in particular to a method and system for constructing a high-speed railway line risk map by considering transportation efficiency loss, which comprises the following steps: constructing a train operation risk event type judgment model, and classifying and identifying station and interval risk events; high-speed rail line transportation efficiency loss is quantified, a calculation model is constructed, risk event probability analysis is carried out based on risk event type judgment and efficiency loss quantification, and space-time grid potential risk influences are measured and calculated; and drawing a line transportation risk map according to the quantized value, and analyzing influence factors. According to the method, various station and interval risk events are effectively extracted and judged, various risk influences are scientifically quantified by innovatively providing the instantaneous transportation loss rate and the line transportation loss rate, accurate mastering of the risk map on the severity level and the influence range of the risk events is improved, and the risk assessment accuracy is improved. Compared with the existing research, the time-space change and distribution condition of the risk influence of each level can be more intuitively mastered.

Description

Technical Field

[0001] The present invention relates to the technical field of transportation engineering, and in particular to a method and system for constructing a high-speed railway line risk map by considering transportation efficiency loss. Background Art

[0002] As my country's high-speed railway transportation development enters a new stage of improving quality and efficiency, building a line risk map is the key foundation for scientifically formulating train operation plans. Faced with risk events such as vehicle failures, track circuit failures, foreign object intrusion, and pantograph failures at stations and sections, how to quantify the impact of different risk events on line train operations and establish a reasonable and reliable line transportation organization risk map has become an important basic work for high-speed railway transportation management departments. Traditional risk warning analysis driven by the statistical probability of risk events has been difficult to adapt to the dynamic and changing transportation organization scenarios and the needs of accurate and efficient train scheduling.

[0003] The identification of high-speed railway train operation risks is a pre-requisite for risk assessment and analysis. At present, the research on high-speed railway transportation risk identification mainly involves risk factor analysis and risk scenario classification. At the risk factor analysis level, existing research has formed a risk factor analysis system covering personnel, equipment, management and environmental risk factors at the levels of railway vehicle maintenance, locomotive maintenance, engineering maintenance, signal, vehicle and passenger transportation. By using methods such as accident tree, Bayesian model and risk propagation chain, the key factors of various types of accidents can be analyzed. At the risk scenario classification level, a large number of studies have established risk scenario clustering models and risk level classification based on historical emergencies and their impact data, considering parameters such as initial delays, number of affected trains and line service frequency, which can describe the transportation characteristics and disturbance characteristics of various types of train operation risk scenarios to a certain extent. However, existing studies focus more on the time and spatial distribution characteristics of various types of risks themselves, and the data statistics and analysis periods span different time nodes. On the one hand, the differences before and after the railway bureau's timetable adjustment nodes are not considered. On the other hand, the risks of various levels and types on the line are not superimposed and considered in a unified time and space perspective. It is difficult to guide the preparation of train operation plans and the optimization of emergency dispatch strategies based on risk identification results.

[0004] At present, the assessment and analysis of high-speed railway train operation risks mainly include risk impact assessment and risk prediction analysis. For risk impact assessment, at the level of line transportation organization, relevant scholars in this field mostly use fuzzy comprehensive evaluation, hierarchical analysis method, and entropy weight method to conduct systematic analysis and risk quantification research. At the event level, they mostly use machine learning methods to analyze the impact of events with scenario parameters and risk factors as input; at the risk prediction analysis level, scholars in this field have combined graph theory, neural network models and other cross-cutting methods to establish relatively mature risk propagation models and state deduction models, which can analyze the impact of various risk events based on established line equipment conditions and transportation plans.

[0005] In summary, the present invention is driven by the train operation data under the disturbance of risk events during holidays. On the one hand, the initial risk disturbance event judgment model and the calculation formula of various risk space-time disturbance probabilities are constructed through the deviation between the actual train operation space-time trajectory and the planned operation diagram; on the other hand, the transportation efficiency loss model is constructed by considering the risk type, disturbance propagation, disturbance recovery and train delay to quantify the impact of single-source risk. Combining the risk probability and impact, a transportation risk calculation model in different space-time grids is proposed, and a multi-source risk distribution map of the line is constructed based on this, and a key factor identification method for high-risk space-time areas is formed. Summary of the invention

[0006] In view of the above problems existing in the prior art, the present invention is proposed.

[0007] Therefore, the technical problem to be solved by the present invention is: to provide a method for constructing a high-speed railway line risk map taking into account the loss of transport efficiency, based on emergency event data, planned operation timetable, actual timetable and train control system monitoring data, at the risk event statistics and analysis level, combined with the train operation plan and the affected situation, the spatiotemporal disturbance probability formulas of various risk events at stations and sections are respectively constructed; at the disturbance impact analysis level, an innovative quantitative model for the loss of transport efficiency of high-speed railway lines is constructed driven by the concept of resilience, and a scientific calibration model and method are established for parameters such as disturbance duration, train grade, risk propagation rate and disturbance recovery rate involved in the quantitative calculation; combined with the spatiotemporal disturbance probability and transport loss impact of various risks, a fine-grained comprehensive line transportation risk measurement model is constructed, risk characteristics and key influencing factors are extracted, and a method for drawing a high-speed railway transportation risk map is formed.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions, which are a method for constructing a high-speed railway line risk map considering the loss of transport efficiency, including: constructing a train operation risk event type determination model; considering the risk impact of line transport efficiency loss, and quantifying the construction of a high-speed railway line transport efficiency loss calculation model; based on the risk event type determination model and the quantitative calculation of line transport efficiency loss, performing risk event probability analysis and measuring the potential risk impact of the space-time grid; based on the quantified value of the risk impact in the space-time grid, drawing a line transport risk map and analyzing the risk impact factors of the risk map data.

[0009] As a preferred solution of the method for constructing a high-speed railway line risk map considering the loss of transportation efficiency described in the present invention, wherein: the risk event type determination model includes an origin delay event determination model, an overtime parking event determination model, a temporary parking event determination model and a speed limit operation event determination model;

[0010] The departure delay event determination model is expressed as:

[0011]

[0012]

[0013] Among them, d w Indicates the initial delay duration caused by the station-originated delay event. represents the actual departure time of train i at station k, represents the scheduled departure time of train i at station k, n represents the number of trains, m represents the number of stations, i, j represents the train number index, and k represents the station number index;

[0014] The first line of constraint condition st indicates the earliest departure train of train i at station k under the influence of risk event. Nan means it does not exist, which means the departure train only has departure time and arrival time. Does not exist, the second line of constraint condition st indicates that the risk event affects the departure delay of the following train i exceeding the 4-min threshold;

[0015] The overtime parking event determination model is expressed as:

[0016]

[0017]

[0018] Among them, d c Indicates the initial delay duration caused by the overtime parking event at the station. Indicates the actual arrival time of the train at the station. Indicates the scheduled arrival time of the train at the station;

[0019] The first line of constraint condition st indicates that the difference between the actual stop time and the planned stop time of train i at station k exceeds the minimum threshold of 4 minutes. The second line of constraint condition st indicates that station k is on the running path of train i. When the planned departure time of the train is equal to the planned arrival time, train i is represented as a through-operation train at station k. When the planned departure time is greater than the stop operation time, it is represented as a stop operation train.

[0020] The temporary parking event determination model is expressed as:

[0021]

[0022]

[0023] Among them, d lt Indicates the initial delay duration caused by the temporary stop event in the section. represents the actual arrival time of train i at station k+1, represents the planned arrival time of train i at station k+1, represents the actual departure time of the qth train after train i at station k, represents the scheduled departure time of the qth train after train i at station k;

[0024] The first row of constraints indicates that the delay time of train i in the interval (k, k+1) satisfies the condition of the shortest temporary stop time. The second row indicates that the running path of train i passes through station k and the interval (k, k+1). The third row indicates that due to the temporary stop of train i in the interval, the first and second subsequent trains are forced to stop overtime at the rear station k.

[0025] The speed limit operation event determination model is expressed as:

[0026]

[0027]

[0028] Among them, d yw Indicates the initial delay duration caused by the interval speed limit operation event;

[0029] The first line of constraints indicates that the operation delay in the interval (k, k+1) meets the judgment threshold of the minimum delay time. The second line of constraints indicates that the train passes through the interval (k, k+1). Train i represents the departure train from station k, and train i represents the arrival train at station k+1.

[0030] As a preferred solution of the method for constructing a high-speed railway line risk map considering the transport efficiency loss of the present invention, wherein: the high-speed railway line transport efficiency loss calculation model includes single train transport efficiency loss rate calculation, line instantaneous transport efficiency loss calculation and line transport efficiency loss morphology analysis and calculation;

[0031] The calculation of the single train transport efficiency loss rate includes the single train instantaneous transport efficiency loss and the single train cumulative transport efficiency loss;

[0032] Among them, when the actual running position of train i is at the station at time t, the instantaneous transport efficiency loss of the train is expressed as:

[0033]

[0034]

[0035] Among them, η i (t) represents the operating efficiency loss of train i at time t, k(t) represents the number of the next train to arrive at the station closest to time t, represents the actual arrival time of train i at station k(t), represents the planned arrival time of train i at station k(t), represents the actual departure time of train i at station k(t), represents the scheduled departure time of train i at the starting station in the section, represents the planned arrival time of train i at the terminal station in the section;

[0036] When the actual running position of train i at time t is within the interval, the delay is estimated by interpolation, which is expressed as:

[0037]

[0038] Among them, delay i (t) represents the actual delay time of train i at time t, represents the delay increment of train i between station k(t) and station k(t)+1 in the actual operation timetable, represents the scheduled departure time of train i at station k(t), represents the actual running time of train i from station k(t) to station k(t)+1;

[0039] The reverse calculation of the planned travel time is expressed as:

[0040]

[0041] in, It represents the planned travel time of train i at time t in the planned operation scheme corresponding to the actual scheme;

[0042] The instantaneous loss rate of a single train is calculated and expressed as:

[0043]

[0044] The cumulative transport efficiency loss of a single train is expressed as:

[0045]

[0046] Among them, η i represents the cumulative transport efficiency loss of train i under the influence of risk events, represents the time when train i starts to be delayed, represents the time when train i leaves the line dispatching area;

[0047] The instantaneous transport efficiency loss calculation of the line includes determining the disturbed train set and considering the instantaneous efficiency loss of the line with train weights;

[0048] The disturbed train set is determined as:

[0049]

[0050] Among them, A T (k) represents the set of disturbed vehicles affected by the risk event r occurring at station k, It represents the scheduled departure time of the affected train number j at station k, Indicates the minimum train departure interval that the station facilities can handle. A represents the average train departure interval at station k during the period when risk event r occurs; r (k, k+1) represents the set of disturbed vehicles affected by the risk event r occurring in the interval (k, k+1), represents the minimum train tracking interval in the interval (k, k+1), represents the average train running interval in the interval (k, k+1) during which the risk event r occurs;

[0051] The instantaneous efficiency loss of the line considering the train weight is expressed as:

[0052]

[0053] Among them, η r (t) represents the instantaneous transport efficiency loss at the line level at time t under the influence of risk event r, A r represents the set of disturbed trains affected by risk event t, w i Represents the weight coefficient of the train, vi represents the travel speed of train i, Indicates the average travel speed of all trains, m i represents the number of train i, L i Represents the length of the running path of train i, m j represents the number of train j, L j represents the running path length of train j;

[0054] The line transport efficiency loss morphological analysis and calculation are expressed as follows:

[0055]

[0056]

[0057]

[0058] Among them, Γ r represents the cumulative transport efficiency loss caused by risk event r on the line, represents the maximum instantaneous transport efficiency loss of the line under the risk event r disturbance, μ r represents the duration of the maximum instantaneous performance loss under the influence of risk event r, Indicates the start time of the maximum instantaneous efficiency loss period, Indicates the end time of the maximum instantaneous efficiency loss period, when the maximum line instantaneous transport efficiency loss Reaching above 95% indicates entering a mid-term stable state.

[0059] As a preferred solution of the method for constructing a high-speed railway line risk map considering the loss of transportation efficiency described in the present invention, wherein: the risk event probability analysis includes station risk event probability analysis, interval risk event probability and the spatiotemporal occurrence probability of the two types of events;

[0060] The probability analysis of the station risk event is expressed as:

[0061]

[0062] in, represents the probability of a station risk event occurring on date type y, y=1 represents holidays, y=2 represents weekdays, y=3 represents weekends, N y (d w ) represents the number of late-starting events corresponding to the statistical date type y, N y (d c ) represents the number of station over-stop events corresponding to the statistical date type y, D y Indicates the number of days of the date type in the historical database, β indicates the date index, and n βIndicates the number of trains running on the line section within the date, path i,k represents a binary variable, 1 means that train number i passes through station k, and 0 means that train number i does not pass through station k. 1 means that train number i passes through station k, and 0 means that train number i does not pass through station k. represents the scheduled departure time of train i at station k on date β, represents the planned arrival time of train i at station k on date β;

[0063] When the train has a departure time at station k, the station path relationship is determined, and when the train has an arrival time at station k, the station path relationship is determined;

[0064] The probability of the interval risk event is expressed as:

[0065]

[0066] in, represents the probability of occurrence of interval risk events under date type y, N y (d lt ) represents the number of temporary parking spaces in the corresponding statistical date type y, N y (d yw ) indicates the number of interval speed limit operation events under the corresponding statistical date type y, represents the binary decision variable of the interval (k, k+1) on the path of train number i, 1 means that train number i passes through the interval, 0 means that train number i does not pass through the interval, represents the scheduled departure time of train i at station k on date β, represents the planned arrival time of train i at station k+1 on date β;

[0067] When the arrival time of the train at the interval station k+1 is greater than the departure time at the station k, it indicates that there is interval operation and the interval path relationship is determined;

[0068] The spatiotemporal occurrence probabilities of the two types of events include the spatiotemporal occurrence probabilities of the departure delay event and the spatiotemporal occurrence probabilities of the interval temporary stop event;

[0069] The spatiotemporal probability of the initial delay event is expressed as:

[0070]

[0071] in, represents the spatiotemporal probability of a delayed event originating at time period pr and position ps under date type y, where pr represents the starting time of the time period, ps represents the number of the initial disturbance position of the risk, P[E(d w )|E(d w ),E(d e)] represents the conditional probability of the initial delay event, Indicates the statistical percentage of departure delays in time period pr under date type y. represents the time distribution function of the initial delay event, (pr,pr+0.5) represents the integration period, It indicates the statistical proportion of delayed events originating at position ps under date type y. Indicates the frequency of the location of the originating delay event, represents the total number of departure delay events occurring at station k under date type y;

[0072] The spatiotemporal probability of the temporary parking event in the section is expressed as:

[0073]

[0074] in, represents the spatiotemporal probability of a temporary stop event in the starting interval at time period pr and location ps under date type y, P[E(d lt )|E(d lt ),E(d yw )] represents the conditional probability of temporary parking event in the starting section, It indicates the statistical proportion of temporary parking time in the starting section within the time period pr under date type y. represents the distribution function of temporary parking events in the originating section, It represents the statistical proportion of temporary parking events in the starting section at position ps under date type y. Indicates the frequency of temporary parking incidents at the originating section. Indicates the number of speed limit operation events that occurred in the interval (k, k+1) under date type y;

[0075] When the position number ps takes the value [1, m-1], there are m-1 intervals in the corresponding line with m stations.

[0076] As a preferred solution of the method for constructing a high-speed railway line risk map considering the transport efficiency loss described in the present invention, wherein: the calculation of the potential risk impact of the space-time grid includes calculating the potential risk impact of the space-time grid on the station risk event and calculating the potential risk impact of the space-time grid on the interval risk event;

[0077] The calculation of the potential risk impact of the space-time grid on the station risk event is expressed as:

[0078]

[0079] in, It represents the quantitative value of the potential risk impact of the station in the spatiotemporal grid (pr,ps) under date type y. represents the average line transport efficiency loss caused by station risk in the space-time network (pr,ps), N y,z (pr,ps) represents the total number of station risk events in the space-time grid under date type y, represents the number of risk events at the rth level of the station in the spatiotemporal grid under date type y, lez represents the level of the risk event at the station, 10 4 represents the year-on-year expansion coefficient;

[0080] The calculation of the potential risk impact of the space-time grid on the interval risk event is expressed as:

[0081]

[0082] in, It represents the quantitative value of the potential risk impact of the interval in the space-time grid (pr,ps) under the date type y. represents the average line transport efficiency loss caused by the risk in the interval of the space-time network (pr, pr), N y,q (pr,ps) represents the total number of risk events in the interval of the space-time grid under date type y, It represents the number of interval risk events of level r in the space-time grid under date type y, and leq represents the level of interval risk events.

[0083] As a preferred solution of the method for constructing a high-speed railway line risk map considering the transport efficiency loss described in the present invention, wherein: the drawing of the line transport risk map includes reading the data of the risk impact time-space distribution matrix, continuously interpolating the line transport risk, and drawing the risk contour lines;

[0084] The risk impact spatiotemporal distribution matrix is ​​expressed as:

[0085]

[0086] Among them, RF y Represents the spatial and temporal distribution matrix of the risk impact of the line under date type y.

[0087] As a preferred solution of the method for constructing a high-speed railway line risk map considering the loss of transportation efficiency described in the present invention, the risk influencing factor analysis of the risk map data includes the construction of a sample attribute set and the analysis of influencing factors based on the fusion of XGBOOST and LightGBM;

[0088] The construction of the sample attribute set includes the analysis of station risk event parameters and the analysis of interval risk event parameters;

[0089] The station risk events include analyzing the quantitative value of station risk impact, the number of one-way arrival and departure lines, the station departure frequency and the station location;

[0090] The section risk events include analyzing the quantified value of the section risk impact, the upper speed limit of the section, the density of train distribution in the section, the section location and the average speed grade coefficient of the vehicle type;

[0091] The influencing factor analysis based on the fusion of XGBOOST and LightGBM includes importing risk factor analysis data, setting XGBOOST training model parameters, training based on XGBOOST, and loading the LightGBM analysis model.

[0092] Another object of the present invention is to provide a high-speed railway line risk map system that takes into account the loss of transport efficiency. By comprehensively analyzing the spatiotemporal distribution of risk events and the loss of transport efficiency, a holographic perception of the spatiotemporal risk points within the line scheduling section can be achieved, thereby improving the reliability of train operation plans.

[0093] In order to solve the above technical problems, the present invention provides the following technical solutions: considering the loss of transportation efficiency, constructing a high-speed railway line risk map system, including: a risk determination model, an efficiency quantification model, a risk probability analysis model and a risk map drawing model;

[0094] The risk determination model determines the types of risk events that occur during high-speed rail operation by identifying and classifying various risk events that affect train operation;

[0095] The efficiency quantification model calculates the specific loss value of high-speed rail line transportation efficiency by quantifying the high-speed rail transportation efficiency loss caused by risk events;

[0096] The risk probability analysis model performs a probability analysis of risk events based on the risk event type determination model and the transport efficiency loss quantification model, and calculates the frequency and probability of occurrence of different types of risk events within the statistical unit;

[0097] The risk map drawing model draws a transportation risk map of the high-speed rail line based on the quantified value of the risk impact in the space-time grid, and conducts detailed data analysis based on the risk influencing factors to identify high-risk areas and time periods.

[0098] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the method for constructing a high-speed railway line risk map considering transportation efficiency loss as described above are implemented.

[0099] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the method for constructing a high-speed railway line risk map taking into account transportation efficiency loss as described above are implemented.

[0100] The beneficial effects of the present invention are as follows: First, based on the planned operation diagram and the actual operation diagram, the risk event information of various stations and sections can be effectively determined and extracted. The present invention abandons the traditional classification method based on the source of risk. Compared with the existing research, it obtains event information from the background of the safety supervision system and performs retrieval, reverse deduction, screening and integration, which is more direct and efficient; second, the risk impact calculation based on operation data is more reliable, focusing on the organic unity of multi-source risk disturbances and their impacts on the train transportation space-time network, innovatively proposing instantaneous transportation loss rate and line transportation loss rate, and summarizing the loss morphological characteristics to scientifically quantify various risk impacts; third, the constructed risk map has a more accurate grasp of the severity level and impact range of line risk events. The algorithm according to the present invention can output a high-precision line risk impact space-time distribution map based on the risk impact numerical matrix and is equipped with risk threshold contour lines to clearly identify high, medium and low risk areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0102] Figure 1 A flow chart of a technical method for constructing a high-speed railway line risk map taking into account transportation efficiency loss, provided in accordance with an embodiment of the present invention.

[0103] Figure 2 A schematic diagram of auxiliary parameters of a model for calculating instantaneous delays of interval trains in a method for constructing a high-speed railway line risk map taking into account transport efficiency losses provided in one embodiment of the present invention.

[0104] Figure 3 A single train instantaneous transport efficiency loss morphological curve and a schematic diagram of calculation parameters for a method for constructing a high-speed railway line risk map taking into account transport efficiency loss provided in one embodiment of the present invention.

[0105] Figure 4 A curve distribution and calculation diagram of line instantaneous transport efficiency loss form 1 of a method for constructing a high-speed railway line risk map taking into account transport efficiency loss provided by an embodiment of the present invention.

[0106] Figure 5 A curve distribution and calculation diagram of form 2 of instantaneous transport efficiency loss of a line in a method for constructing a high-speed railway line risk map taking into account transport efficiency loss provided by an embodiment of the present invention.

[0107] Figure 6A train operation diagram of the Xuzhou East-Nanjing South line section of the Beijing-Shanghai High-Speed ​​Railway, which provides a method for constructing a high-speed railway line risk map taking into account transportation efficiency loss, according to an embodiment of the present invention.

[0108] Figure 7 A transport efficiency loss curve diagram of affected trains and lines under a typical departure delay risk event for a method for constructing a high-speed railway line risk map taking into account transport efficiency loss provided by an embodiment of the present invention.

[0109] Figure 8 A transport efficiency loss curve diagram of affected trains and lines under a typical section temporary parking risk event in a method for constructing a high-speed railway line risk map considering transport efficiency loss provided by an embodiment of the present invention.

[0110] Fig. 9 The invention provides a method for constructing a high-speed railway line risk map taking into account the loss of transportation efficiency, which is output by an algorithm of the line risk threshold contour lines and the risk spatiotemporal distribution map.

[0111] Fig.10 A fitting validity diagram of typical station risk scenarios based on XGBoost training for a method for constructing a high-speed railway line risk map taking into account transportation efficiency loss provided by an embodiment of the present invention.

[0112] Fig.11 A graph showing the results of risk factor analysis of typical stations based on LightGBM regression for a method for constructing a high-speed railway line risk map taking into account transport efficiency loss provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0113] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0114] Example 1, reference Figure 1-Figure 5 , is an embodiment of the present invention, which provides a method for constructing a high-speed railway line risk map considering transportation efficiency loss, including:

[0115] S1: Construct a model for determining the type of train operation risk events.

[0116] It should be noted that risk event types are divided into station risk events and section risk events, specifically into delayed departure events, overtime stop events, temporary stop events and speed limit operation events.

[0117] Furthermore, the risk event type determination model includes an origin delay event determination model, an overtime parking event determination model, a temporary parking event determination model, and a speed limit operation event determination model;

[0118] According to the statistics of the background safety monitoring record data, the departure delays of 3 minutes or less at stations account for 93% of the global delay events, which can be regarded as an acceptable timetable operation deviation. Therefore, the present invention takes 4 minutes as the minimum threshold for the risk disturbance to cause departure delays. The departure delay event judgment model is expressed as:

[0119]

[0120]

[0121] Among them, d w It indicates the initial delay duration caused by the station departure delay event, which is the maximum value of all the delays of the delayed departure trains under the influence of the risk. represents the actual departure time of train i at station k, represents the scheduled departure time of train i at station k, n represents the number of trains, m represents the number of stations, i, j represents the train number index, and k represents the station number index;

[0122] The first line of constraint condition st indicates the earliest departure train of train i at station k under the influence of risk event. Nan means it does not exist, which means the departure train only has departure time and arrival time. Does not exist, the second line of constraint condition st indicates that the risk event affects the departure delay of the following train i exceeding the 4-min threshold;

[0123] Taking 4 minutes as the minimum judgment threshold of overtime parking events, the overtime parking event judgment model is expressed as:

[0124]

[0125]

[0126] Among them, d c It indicates the initial delay duration caused by the overtime stop event at the station, which is the maximum overtime stop duration of the affected trains under the risk influence. Indicates the actual arrival time of the train at the station. Indicates the scheduled arrival time of the train at the station;

[0127] The first line of constraint condition st indicates that the difference between the actual stop time and the planned stop time of train i at station k exceeds the minimum threshold of 4 minutes. The second line of constraint condition st indicates that station k is on the running path of train i. When the planned departure time of the train is equal to the planned arrival time, train i is represented as a through-operation train at station k. When the planned departure time is greater than the stop operation time, it is represented as a stop operation train.

[0128] Statistics show that it takes at least 5 minutes for the on-board mechanic to get off the vehicle for inspection. Considering that the loss time of vehicle braking and acceleration is at least 5 minutes, the temporary parking event judgment model is expressed as:

[0129]

[0130]

[0131] Among them, d lt Indicates the initial delay duration caused by the temporary stop event in the section. represents the actual arrival time of train i at station k+1, represents the planned arrival time of train i at station k+1, represents the actual departure time of the qth train after train i at station k, represents the scheduled departure time of the qth train after train i at station k;

[0132] The first row of constraints indicates that the delay time of train i in the interval (k, k+1) satisfies the condition of the shortest temporary stop time. The second row indicates that the running path of train i passes through station k and the interval (k, k+1). The third row indicates that due to the temporary stop of train i in the interval, the first and second subsequent trains are forced to stop overtime at the rear station k.

[0133] The speed limit operation event determination model generally adopts two schemes, including location speed limit and vehicle speed limit;

[0134] The shortest speed limit is 4km before and after the accident site, with a speed limit of less than 160km / h. Since different models have different speeds in the interval, the minimum interval delay caused by their speed limit operation is also different. The speed limit is expressed as:

[0135]

[0136]

[0137]

[0138] in, It indicates the delay time of the shortest section of train i caused by the speed limit at the location. It represents the braking loss time caused by speed limit operation of train i in the common braking mode. represents the starting loss time caused by speed limit operation of train i in the common braking mode, v i Indicates the speed of the train in the corresponding section, d yw Indicates the initial delay duration caused by the interval speed limit operation event, which is the maximum value of the interval delay of the speed limit operation train in the corresponding period;

[0139] The first line of constraints indicates that the operation delay in the interval (k, k+1) meets the judgment threshold of the minimum delay time. The second line of constraints indicates that the train passes through the interval (k, k+1). Train i represents the departure train from station k, and train i represents the arrival train at station k+1.

[0140] There is no unified standard for vehicle speed limit. Under risk disturbance, the train can run at the speed limit for 5 minutes or run at the speed limit to the next station. The minimum interval operation delay caused by vehicle speed limit can be regarded as equivalent to the minimum interval operation delay caused by location speed limit. Since both location speed limit and vehicle speed limit belong to the interval speed limit operation event type, they can be judged according to the same model.

[0141] The loss time and minimum delay time under three operating speeds are accurately calculated as shown in Table 1;

[0142] Table 1 Acceleration and deceleration loss time of representative high-speed train models under the 160km / h speed limit

[0143]

[0144] Furthermore, the risk event type determination mechanism is completed through 5 steps:

[0145] (1) Schedule deviation calculation;

[0146] (2) Considering the six-hour skylight window, the detection time domain covers 6:00 a.m. to 24:00 a.m., and the whole day is divided into 36 time periods for traversal retrieval with a statistical step length of 30 minutes;

[0147] (3) Determination of station risk events;

[0148] (4) Determination of interval risk events;

[0149] (5) Determination of interval risk events.

[0150] S2: Considering the risk impact of line transport efficiency loss, a quantitative calculation model for high-speed rail line transport efficiency loss is constructed.

[0151] It should be noted that the high-speed rail line transport efficiency loss calculation model includes the calculation of single train transport efficiency loss rate, the calculation of line instantaneous transport efficiency loss, and the analysis and calculation of line transport efficiency loss morphology;

[0152] The calculation of the single train transport efficiency loss rate includes the single train instantaneous transport efficiency loss and the single train cumulative transport efficiency loss;

[0153] Among them, when the actual running position of train i is at the station at time t, the instantaneous transport efficiency loss of the train is expressed as:

[0154]

[0155]

[0156] Among them, η i (t) represents the operating efficiency loss of train i at time t, k(t) represents the number of the next train to arrive at the station closest to time t, represents the actual arrival time of train i at station k(t), represents the planned arrival time of train i at station k(t), represents the actual departure time of train i at station k(t), represents the scheduled departure time of train i at the starting station in the section, It represents the planned arrival time of train i at the terminal station in the section. The related auxiliary calculation parameters are shown as follows: Figure 2 As shown;

[0157] When the actual running position of train i at time t is within the interval, the delay is estimated by interpolation, which is expressed as:

[0158]

[0159] Among them, delay i (t) represents the actual delay time of train i at time t, represents the delay increment of train i between station k(t) and station k(t)+1 in the actual operation timetable, represents the scheduled departure time of train i at station k(t), represents the actual running time of train i from station k(t) to station k(t)+1;

[0160] The reverse calculation of the planned travel time is expressed as:

[0161]

[0162] in, It represents the planned travel time of train i at time t in the planned operation scheme corresponding to the actual scheme;

[0163] The instantaneous loss rate of a single train is calculated and expressed as:

[0164]

[0165] like Figure 3 As shown in the figure, according to the analysis of previous historical data, there are two main forms of the change curve of the instantaneous transport efficiency loss rate of a single train: in form 1, affected by the train operation priority and line transport service frequency, the train cannot run on time and needs to avoid other trains, resulting in a gradual increase in the instantaneous transport capacity loss rate; in form 2, due to the relatively loose timetable redundancy, the train can run on time by increasing the section speed and reducing the stop time; the cumulative transport efficiency loss of a single train is expressed as:

[0166]

[0167] Among them, η i represents the cumulative transport efficiency loss of train i under the influence of risk events, represents the time when train i starts to be delayed, represents the time when train i leaves the line dispatching area;

[0168] The calculation of line instantaneous transport efficiency loss includes the determination of disturbed train sets and the line instantaneous efficiency loss considering train weights;

[0169] The disturbed train set is determined as:

[0170]

[0171] Among them, A T (k) represents the set of disturbed vehicles affected by the risk event r occurring at station k, It represents the scheduled departure time of the affected train number j at station k, Indicates the minimum train departure interval that the station facilities can handle. A represents the average train departure interval at station k during the period when risk event r occurs; r (k, k+1) represents the set of disturbed vehicles affected by the risk event r occurring in the interval (k, k+1), represents the minimum train tracking interval in the interval (k, k+1), represents the average train running interval in the interval (k, k+1) during which the risk event r occurs;

[0172] The instantaneous efficiency loss of the line considering the train weight is expressed as:

[0173]

[0174] Among them, η r (t) represents the instantaneous transport efficiency loss at the line level at time t under the influence of risk event r, A r represents the set of disturbed trains affected by risk event t, w i Represents the weight coefficient of the train, v i represents the travel speed of train i, Indicates the average travel speed of all trains, m i represents the number of train i, L i Represents the length of the running path of train i, m j represents the number of train j, L j represents the running path length of train j;

[0175] Based on the analysis of historical risk data, the present invention summarizes two forms of line transportation efficiency loss curves. Form 1 is as follows: Figure 4 As shown, form 2 is Figure 5 As shown;

[0176] Figure 4 : Form 1a: It shows a trend of rapid rise in the early stage and slow decline in the later stage. This form is often seen in risk disturbance scenarios in line sections with serious initial delays and busy transportation;

[0177] Form 1b: It shows a trend of rapid rise in the early stage and rapid decline in the later stage. This form is often seen in risk disturbance scenarios where the initial delay is serious but the service frequency of the line section has a large redundancy;

[0178] Form 1c: It shows a trend of slow increase in the early stage and rapid decrease in the later stage. This form is often seen in risk disturbance scenarios with short initial delay or low line transportation frequency in the section, that is, the risk disturbance spreads slowly between trains and the affected trains can recover quickly.

[0179] Figure 5 Form 2a: It shows a trend of rapid rise in the early stage, stability in the middle stage, and slow decline in the late stage. This form is often seen in risk disturbance scenarios with serious initial delays, high transportation frequency of the line, and the initial impact position close to the starting point of the dispatching section;

[0180] Form 2b: It shows a trend of rapid rise in the early stage, stability in the middle stage, and rapid decline in the later stage. This form is often seen in risk disturbance scenarios where the initial delay is moderate, the service frequency of the line section is high, and the initial impact position is located in the second half of the dispatching section;

[0181] The analysis and calculation of line transport efficiency loss is expressed as follows:

[0182]

[0183]

[0184]

[0185] Among them, Γ r represents the cumulative transport efficiency loss caused by risk event r on the line, represents the maximum instantaneous transport efficiency loss of the line under the risk event r disturbance, μ r represents the duration of the maximum instantaneous performance loss under the influence of risk event r, Indicates the start time of the maximum instantaneous efficiency loss period, Indicates the end time of the maximum instantaneous efficiency loss period, when the maximum line instantaneous transport efficiency loss Reaching above 95% indicates entering a mid-term stable state.

[0186] S3: Based on the risk event type determination model and quantitative calculation of line transport efficiency loss, conduct risk event probability analysis and measure the potential risk impact of space-time grids.

[0187] It should be noted that the risk event probability analysis includes station risk event probability analysis, interval risk event probability analysis, and the spatiotemporal probability of the two types of events;

[0188] The probability analysis of station risk events is expressed as:

[0189]

[0190] in, represents the probability of a station risk event occurring on date type y, y=1 represents holidays, y=2 represents weekdays, y=3 represents weekends, N y (d w ) represents the number of late-starting events corresponding to the statistical date type y, N y (d c ) represents the number of station over-stop events corresponding to the statistical date type y, D y Indicates the number of days of the date type in the historical database, β indicates the date index, and n β Indicates the number of trains running on the line section within the date, path i,k Indicates whether station k is on the route of train i. 1 means train i passes through station k, and 0 means train i does not pass through station k. represents the scheduled departure time of train i at station k on date β, represents the planned arrival time of train i at station k on date β;

[0191] Taking into account the existence of departing trains (there is no arrival time at the starting station), cross-section running trains (there is no departure time at the terminal station) and section running trains (there is no arrival and departure time at stations outside the section), the station path relationship can be determined when the train has a departure time or arrival time at the station.

[0192] When the train has a departure time at station k, the station path relationship is determined, and when the train has an arrival time at station k, the station path relationship is determined;

[0193] The probability of interval risk events is expressed as:

[0194]

[0195] in, represents the probability of occurrence of interval risk events under date type y, N y (d lt ) represents the number of temporary parking spaces in the corresponding statistical date type y, N y (d yw ) indicates the number of interval speed limit operation events under the corresponding statistical date type y, represents the binary decision variable of the interval (k, k+1) on the path of train number i, 1 means that train number i passes through the interval, 0 means that train number i does not pass through the interval, represents the scheduled departure time of train i at station k on date β, represents the planned arrival time of train i at station k+1 on date β;

[0196] When the arrival time of the train at the interval station k+1 is greater than the departure time at the station k, it indicates that there is interval operation and the interval path relationship is determined;

[0197] The spatiotemporal probability of the two types of events include the spatiotemporal probability of the departure delay event and the spatiotemporal probability of the interval temporary stop event;

[0198] The spatiotemporal probability of the initial delay event is expressed as:

[0199]

[0200] in, represents the spatiotemporal probability of a delayed event originating at time period pr and location ps under date type y, where pr represents the starting time of the time period, and ps represents the location number of the initial disturbance of the risk. For station-type risk events, its value ranges from 1 to the upper limit of the number of stations m. P[E(d w )|E(d w ),E(d e)] represents the conditional probability of the departure delay event, which specifically represents the probability of the station departure delay event under the condition of the station risk event. It is calculated based on the frequency of the two types of events: departure delay and overtime parking. Indicates the statistical percentage of departure delays in time period pr under date type y. represents the time distribution function of the initial delay event, (pr,pr+0.5) represents the integration period, It indicates the statistical proportion of delayed events originating at position ps under date type y. Indicates the frequency of the location of the originating delay event, represents the total number of departure delay events occurring at station k under date type y;

[0201] The spatiotemporal probability of a temporary stop event in a section is expressed as:

[0202]

[0203]

[0204] in, represents the spatiotemporal probability of a temporary stop event in the starting interval at time period pr and location ps under date type y, P[E(d lt )|F(d lt ),E(d yw )] represents the conditional probability of temporary parking event in the starting section, It indicates the statistical proportion of temporary parking time in the starting section within the time period pr under date type y. represents the distribution function of temporary parking events in the originating section, It represents the statistical proportion of temporary parking events in the starting section at position ps under date type y. Indicates the frequency of temporary parking incidents at the originating section. It indicates the number of speed limit operation events in the interval (k, k+1) under date type y. When the position number ps takes the value [1, m-1], there are m-1 intervals in the corresponding line m stations.

[0205] Furthermore, the calculation of the potential risk impact of the space-time grid includes the calculation of the potential risk impact of the space-time grid on the station risk events and the calculation of the potential risk impact of the space-time grid on the interval risk events;

[0206] The calculation of the potential risk impact of the space-time grid on the station risk event is expressed as:

[0207]

[0208]

[0209] in, It represents the quantitative value of the potential risk impact of the station in the spatiotemporal grid (pr,ps) under date type y. represents the average line transport efficiency loss caused by each level of station risk in the space-time network (pr,ps), N y,z (pr,ps) represents the total number of station risk events in the space-time grid under date type y, represents the number of risk events at the rth level of the station in the spatiotemporal grid under date type y, lez represents the level of the risk event at the station, 10 4 represents the year-on-year expansion coefficient;

[0210] The calculation of the potential risk impact of the space-time grid on the interval risk event is expressed as:

[0211]

[0212]

[0213] in, It represents the quantitative value of the potential risk impact of the interval in the space-time grid (pr,ps) under the date type y. represents the average line transport efficiency loss caused by the risk within the interval of the space-time network (pr,ps), N y,q (pr,ps) represents the total number of risk events in the interval of the space-time grid under date type y, It represents the number of interval risk events of level r in the space-time grid under date type y, and leq represents the level of interval risk events.

[0214] S4: Based on the quantified values ​​of risk impact in the space-time grid, draw a line transportation risk map and analyze the risk impact factors of the risk map data.

[0215] It should be noted that drawing the route transport risk map includes reading the data of the risk impact spatiotemporal distribution matrix, continuously interpolating the route transport risk, and drawing the risk contour lines. The specific Python implementation code is shown in Table 2;

[0216] The risk impact spatiotemporal distribution matrix is ​​expressed as:

[0217]

[0218] Among them, RF y Represents the spatial and temporal distribution matrix of the risk impact of the line under date type y.

[0219] Table 2 Risk map drawing main program code

[0220]

[0221]

[0222] Furthermore, the risk influencing factor analysis of the risk map data includes the construction of the sample attribute set and the influencing factor analysis based on the fusion of XGBOOST and LightGBM. The python code written by the present invention is shown in Table 3;

[0223] The construction of sample attribute set includes the analysis of station risk event parameters and interval risk event parameters;

[0224] For station risk events, the analysis includes the quantitative value of station risk impact, the number of one-way arrival and departure lines, the station departure frequency and the station location;

[0225] Among them, the quantitative value of station risk impact is a decision-making attribute, and the risk value of the station at the corresponding time and space coordinate position in the risk map is extracted; the number of one-way arrival and departure lines is a conditional attribute. According to the plan layout of each station on the line, the number of tracks used by the station for receiving and dispatching trains in the current train operation direction is extracted. This value will affect the initial disturbance duration of the station's overtime parking event to a certain extent; the station departure frequency is a conditional attribute, and the departure frequency of the station within 1 hour forward of the corresponding time coordinate is extracted. The higher the value, the greater the traffic density in the line section in front of the station, and the greater the line transportation efficiency loss caused by the departure delay or temporary parking event; the station location is a conditional attribute. According to the line transportation efficiency loss form, it can be seen that the corresponding loss form and cumulative loss under risk events at different locations are different.

[0226] For interval risk events, the analysis includes the quantitative value of interval risk impact, the upper speed limit of the interval, the density of train distribution in the interval, the interval location and the average speed grade coefficient of the vehicle type;

[0227] Among them, the interval risk impact quantification value is a decision attribute, and the interval risk value at the corresponding space-time coordinate position in the risk map is extracted; the interval speed limit upper limit is a conditional attribute, and the speed limit upper limit value of the current interval is extracted according to the line design data. This value is affected by the line geometry parameters such as line shape, curvature, slope, etc., and will affect the transportation efficiency loss to a certain extent, and is indirectly related to the risk impact value; the interval train distribution density is a conditional attribute, and the average running distance of trains passing through the current interval position within 1 hour is extracted. The smaller the value, the shorter the distance between adjacent trains, and the more likely it is to have mutual influence and cause risk diffusion; the interval position is Conditional attributes. According to the form of line transport efficiency loss, the corresponding loss form and cumulative loss under risk events at different locations are different. The average speed grade coefficient of vehicle type is a conditional attribute. Since different types of trains have different maximum operating speeds, their sensitivity to risks is also different, that is, under the current emergency dispatch strategy mechanism, the impact of risks on trains with different operating speeds is different. The present invention defines the speed grade coefficient of interval vehicle type as the grade average of trains passing through the interval within 1 hour, where the grade coefficients of trains with maximum operating speeds of 350, 310 and 250 are 1.4, 1.2 and 1.0 respectively.

[0228] The influencing factor analysis based on the fusion of XGBOOST and LightGBM includes: importing risk factor analysis data, with the characteristic variables being the conditional attributes in the risk sample set as the characteristic variables of the training model, and the decision attributes being the target variables of the training model; setting the XGBOOST training model parameters, setting the training optimization target to variance, using a binary tree as the accelerator, with the initial maximum subtree depth of 5 layers, and the proportion of training samples set according to the sample size; training based on XGBOOST, calculating the mean square error and goodness of fit, and when the goodness of fit exceeds 0.5, the model is considered to be effective and can basically capture the impact of different conditional attributes on risk quantification, otherwise the model training parameters are readjusted; loading the LightGBM analysis model, and outputting the importance of different conditional attributes under various risk scenarios.

[0229] Table 3 XGBOOST+LightGBM fusion programming code

[0230]

[0231]

[0232]

[0233] The above is a schematic scheme of the method for constructing a high-speed railway line risk map considering the loss of transportation efficiency in this embodiment. It should be noted that the technical scheme of the system for constructing a high-speed railway line risk map considering the loss of transportation efficiency and the technical scheme of the method for constructing a high-speed railway line risk map considering the loss of transportation efficiency belong to the same concept. The details not described in detail in the technical scheme of the system for constructing a high-speed railway line risk map considering the loss of transportation efficiency in this embodiment can be referred to the description of the technical scheme of the method for constructing a high-speed railway line risk map considering the loss of transportation efficiency.

[0234] In this embodiment, a high-speed railway line risk map system is constructed by considering the loss of transportation efficiency, which is characterized by: including: a risk determination model, an efficiency quantification model, a risk probability analysis model and a risk map drawing model;

[0235] The risk determination model determines the types of risk events that occur during high-speed rail operation by identifying and classifying various risk events that affect train operation;

[0236] The efficiency quantification model calculates the specific loss value of high-speed rail line transportation efficiency by quantifying the high-speed rail transportation efficiency loss caused by risk events;

[0237] The risk probability analysis model performs a probability analysis of risk events based on the risk event type determination model and the transport efficiency loss quantification model, and calculates the frequency and probability of occurrence of different types of risk events within the statistical unit;

[0238] The risk map drawing model draws a transportation risk map of the high-speed rail line based on the quantified value of the risk impact in the space-time grid, and conducts detailed data analysis based on the risk influencing factors to identify high-risk areas and time periods.

[0239] This embodiment also provides a computing device, which is applicable to the method of constructing a high-speed railway line risk map considering transportation efficiency loss, and includes:

[0240] A memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method of constructing a high-speed railway line risk map taking into account transportation efficiency loss as proposed in the above embodiment.

[0241] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for constructing a high-speed railway line risk map considering transportation efficiency loss as proposed in the above embodiment is implemented.

[0242] The storage medium proposed in this embodiment and the method for constructing a high-speed railway line risk map considering transportation efficiency loss proposed in the above embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0243] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0244] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0245] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0246] Example 2, reference Figure 3-Figure 11, which is an embodiment of the present invention, provides a method for constructing a high-speed railway line risk map taking into account the loss of transportation efficiency. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0247] The present invention selects the working days (a total of 10 days of data) of the Xuzhou East-Nanjing South dispatching section of the Beijing-Shanghai High-speed Railway as the object to illustrate the specific implementation method. The actual background data obtained from the Shanghai Railway Bureau is as follows Figure 6 As shown. There are 6 stations (m=6) that handle Beijing-Shanghai high-speed railway train operations in the current line transportation section, namely: Xuzhou East Station, Suzhou East Station, Bengbu South Station, Dingyuan Station, Chuzhou Station, Nanjing South Station. Among them, Xuzhou East Station not only handles Beijing-Shanghai high-speed railway trains that stop at the station, but also handles the departure trains of this station; Bengbu South Station not only handles the passing and stop operations of trains coming from Xuzhou in the current dispatching section, but also handles a small number of departure trains. An average of 101 trains run from Xuzhou East to Nanjing South every day, and an average of 4 trains run from Bengbu South to Nanjing South every day.

[0248] (1) Risk event determination and statistics:

[0249] According to the present invention, S1 determines and counts the occurrence frequency of various events within the research date period, and the statistical results of various events are shown in Table 4 and Table 5.

[0250] Table 4 Statistical results of station risk event determination (frequency)

[0251]

[0252]

[0253] Table 5 Statistical results of interval risk event determination (frequency)

[0254]

[0255]

[0256] (2) Quantitative calculation of line transport efficiency loss:

[0257] Since risks at different locations and times have different disruptive effects on line transportation, typical station and section risk events are selected here to illustrate the quantitative calculation process of transportation efficiency loss.

[0258] Station risk event: This event corresponds to the first departure delay risk event that occurred at Xuzhou East Station in the third time segment (7:00-7:30) in Table 4. According to the operation diagram, the event lasted 21 minutes, causing delays in two subsequent departure trains and two stopover trains. The transportation efficiency losses of each train in the actual operation diagram all meet the requirements. Figure 3 The curve of form 2, the basic properties of the vehicle and the instantaneous performance loss of the key position nodes are shown in Table 6.

[0259] Table 6 Affected train parameters and delays under typical station risk event scenarios

[0260]

[0261] According to the change of instantaneous transport efficiency loss of each train, the change curve of line instantaneous transport efficiency loss rate can be obtained by combining the formula, such as Figure 7 The calculation of the line transport efficiency loss rate of the present invention takes into account the train weight coefficient. The change form of the line transport loss considering the weight in the current scenario is: Figure 4 The corresponding form 1a in is equal to 0.396, and They are 7.0h and 9.92h (9:55) respectively. Substituting them into the formula, we can get the estimated value of the line transportation efficiency loss caused by the current departure delay event is 0.578.

[0262] Section risk event: This event corresponds to the section temporary parking risk event that occurred in the section between Bengbu South and Dingyuan Station in the 7th time segment (9:00-9:30) in Table 5. According to the operation diagram, this event caused the first G7619 train to be delayed for 19 minutes in the section, and caused the subsequent three trains to have different degrees of departure delays and arrival delays at the stations before and after the section. The single train transportation efficiency loss of each train in the actual operation diagram also meets the requirements. Figure 3 The curve of form 2, the basic properties of the vehicle and the instantaneous performance loss of the key position nodes are shown in Table 7.

[0263] Table 7 Affected train parameters and delays under typical interval risk event scenarios

[0264]

[0265] According to the change of instantaneous transport efficiency loss of each train, the change curve of line instantaneous transport efficiency loss rate can be obtained by combining the formula, such as Figure 8 The calculation of the line transport efficiency loss rate of the present invention takes into account the train weight coefficient. The change form of the line transport loss considering the weight in the current section temporary parking risk disturbance scenario is: Figure 5 The form 2a in is equal to 0.255, and They are 9.25h and 10.9h respectively, and the duration of the loss peak period is μ r It is 0.4h. Substituting it into the formula, we can get the estimated value of the line transportation efficiency loss caused by the current departure delay event as 0.261.

[0266] (3) Mapping of route transport risks:

[0267] Based on the statistical data in Table 4 and Table 5, the line transport loss rate under the influence of individual risk disturbances is obtained according to S2. The two are combined and substituted into the formula to calculate the risk impact quantification value of each space-time grid. Based on this, the risk impact space-time distribution matrix in the form of a formula is obtained, as shown in Table 8. The position numbers 1, 3, 5, 7, 9, and 11 in the figure correspond to the 6 stations in the current dispatching section, and the numbers 2, 4, 6, 8, and 10 correspond to the 5 station intervals. Among them, the highest risk space-time grid is the time period 7.5-8 (7:30-8:00) and position 3 (Suzhou East Station), and the corresponding risk quantification value is 4.34.

[0268] Table 8 Risk impact numerical matrix within the line section

[0269]

[0270]

[0271]

[0272] Based on the risk impact numerical matrix in Table 8, loading the drawing code written in S4-2 of the present invention, we can get Fig. 9 The risk impact contours and risk area maps are shown in Figure 1. The output risk map shows that there are 3 high-risk areas, 5 medium-risk areas and several slight-risk areas in the operating section of the Xuzhou East-Nanjing South high-speed railway line based on the existing data. The 3 high-risk time and space areas are: ①6:15-9:00, Xuzhou East Station; ②7:00-8:30, Suzhou East Station + Suzhou East-Bengbu South Section; ③11:00-12:00, Suzhou East Station. The 5 medium-risk time and space areas are: ①10:10-10:50, Bengbu South Station; ②13:00-14:00, Xuzhou East Station; ③18:00-17:00, Xuzhou East Station; ④10:30-11:00, Dingyuan-Chuzhou Section; ⑤16:30-17:30, Dingyuan Station.

[0273] (4) Risk impact parameter analysis based on risk map:

[0274] Taking station risk events as an example, a station risk sample attribute set is constructed based on risk map data according to S4-3, as shown in Table 9.

[0275] Table 9 Station risk event sample attribute set

[0276]

[0277] Based on the attribute set shown in the table above, XGBoost is used to test whether the current sample data set can be applied to the reinforcement learning algorithm. After debugging, under the current sample conditions, when the training parameters test_size = 0.25, random_state = 5, and max_depth = 4, it has a good regression explanation effect. The corresponding prediction value mean square error is 0.0792, the R square value is 0.5147, and the correlation coefficient is about 0.72, which has good applicability. Therefore, LightGBM is further used for importance analysis. For station-type risk events, the importance of the three conditional parameters is from high to low: departure frequency, station location, and number of arrival and departure lines. The results of XGBoost test and LightGBM analysis are shown in the figure below. Fig.10 , Fig.11 Therefore, in the process of preparing or adjusting the train operation plan for subsequent line sections, various risk event-related factors can be considered for adjustment. For example, for the risks of departure delays and excessive stops at stations within the current section, the departure frequency of some stations upstream of the section can be reduced by adjusting the stop time, interval operation speed and other strategies.

[0278] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for constructing a high-speed railway line risk map considering transportation efficiency loss, characterized by: include: Construct a model for determining the type of train operation risk events; Considering the risk impact of line transport efficiency loss, a quantitative calculation model for high-speed rail line transport efficiency loss is constructed; Based on the risk event type determination model and the quantitative calculation of line transportation efficiency loss, the risk event probability analysis is carried out to measure the potential risk impact of the space-time grid; Based on the quantified values ​​of risk impact in the space-time grid, a line transportation risk map is drawn, and the risk impact factor analysis of the risk map data is carried out.

2. The method for constructing a high-speed railway line risk map considering transportation efficiency loss according to claim 1, characterized in that: The risk event type determination model includes an origin delay event determination model, an overtime parking event determination model, a temporary parking event determination model, and a speed limit operation event determination model; The departure delay event determination model is expressed as: Among them, d w Indicates the initial delay duration caused by the station-originated delay event. represents the actual departure time of train i at station k, represents the scheduled departure time of train i at station k, n represents the number of trains, m represents the number of stations, i, j represents the train number index, and k represents the station number index; The first line of constraint condition st indicates the earliest departure train of train i at station k under the influence of risk event. Nan means it does not exist, which means the departure train only has departure time and arrival time. Does not exist, the second line of constraint condition st indicates that the risk event affects the departure delay of the following train i exceeding the 4-min threshold; The overtime parking event determination model is expressed as: Among them, d c Indicates the initial delay duration caused by the overtime parking event at the station. Indicates the actual arrival time of the train at the station. Indicates the scheduled arrival time of the train at the station; The first line of constraint condition st indicates that the difference between the actual stop time and the planned stop time of train i at station k exceeds the minimum threshold of 4 minutes. The second line of constraint condition st indicates that station k is on the running path of train i. When the planned departure time of the train is equal to the planned arrival time, train i is represented as a through-operation train at station k. When the planned departure time is greater than the stop operation time, it is represented as a stop operation train. The temporary parking event determination model is expressed as: Among them, d lt Indicates the initial delay duration caused by the temporary stop event in the section. represents the actual arrival time of train i at station k+1, represents the planned arrival time of train i at station k+1, represents the actual departure time of the qth train after train i at station k, represents the scheduled departure time of the qth train after train i at station k; The first row of constraints indicates that the delay time of train i in the interval (k, k+1) satisfies the condition of the shortest temporary stop time. The second row indicates that the running path of train i passes through station k and the interval (k, k+1). The third row indicates that due to the temporary stop of train i in the interval, the first and second subsequent trains are forced to stop overtime at the rear station k. The speed limit operation event determination model is expressed as: Among them, d yw Indicates the initial delay duration caused by the interval speed limit operation event; The first line of constraints indicates that the operation delay in the interval (k, k+1) meets the judgment threshold of the minimum delay time. The second line of constraints indicates that the train passes through the interval (k, k+1). Train i represents the departure train from station k, and train i represents the arrival train at station k+1.

3. The method for constructing a high-speed railway line risk map considering transportation efficiency loss according to claim 2, characterized in that: The high-speed railway line transportation efficiency loss calculation model includes single train transportation efficiency loss rate calculation, line instantaneous transportation efficiency loss calculation and line transportation efficiency loss form analysis and calculation; The calculation of the single train transport efficiency loss rate includes the single train instantaneous transport efficiency loss and the single train cumulative transport efficiency loss; Among them, when the actual running position of train i is at the station at time t, the instantaneous transport efficiency loss of the train is expressed as: Among them, η i (t) represents the operating efficiency loss of train i at time t, k(t) represents the number of the next train to arrive at the station closest to time t, represents the actual arrival time of train i at station k(t), represents the planned arrival time of train i at station k(t), represents the actual departure time of train i at station k(t), represents the scheduled departure time of train i at the starting station in the section, represents the planned arrival time of train i at the terminal station in the section; When the actual running position of train i at time t is within the interval, the delay is estimated by interpolation, which is expressed as: Among them, delay i (t) represents the actual delay time of train i at time t, represents the delay increment of train i between station k(t) and station k(t)+1 in the actual operation timetable, represents the scheduled departure time of train i at station k(t), represents the actual running time of train i from station k(t) to station k(t)+1; The reverse calculation of the planned travel time is expressed as: in, It represents the planned travel time of train i at time t in the planned operation scheme corresponding to the actual scheme; The instantaneous loss rate of a single train is calculated and expressed as: The cumulative transport efficiency loss of a single train is expressed as: Among them, η i represents the cumulative transport efficiency loss of train i under the influence of risk events, represents the time when train i starts to be delayed, represents the time when train i leaves the line dispatching area; The instantaneous transport efficiency loss calculation of the line includes determining the disturbed train set and considering the instantaneous efficiency loss of the line with train weights; The disturbed train set is determined as: Among them, A T (k) represents the set of disturbed vehicles affected by the risk event r occurring at station k, It represents the scheduled departure time of the affected train number j at station k, Indicates the minimum train departure interval that the station facilities can handle. A represents the average train departure interval at station k during the period when risk event r occurs; r (k, k+1) represents the set of disturbed vehicles affected by the risk event r occurring in the interval (k, k+1), represents the minimum train tracking interval in the interval (k, k+1), represents the average train running interval in the interval (k, k+1) during which the risk event r occurs; The instantaneous efficiency loss of the line considering the train weight is expressed as: Among them, η r (t) represents the instantaneous transport efficiency loss at the line level at time t under the influence of risk event r, A r represents the set of disturbed trains affected by risk event t, w i Represents the weight coefficient of the train, v i represents the travel speed of train i, Indicates the average travel speed of all trains, m i represents the number of train i, L i Represents the length of the running path of train i, m j represents the number of train j, L j represents the running path length of train j; The line transport efficiency loss morphological analysis and calculation are expressed as follows: Among them, Γ r represents the cumulative transport efficiency loss caused by risk event r on the line, represents the maximum instantaneous transport efficiency loss of the line under the disturbance of risk event r, μr represents the duration of the maximum instantaneous efficiency loss under the influence of risk event r, Indicates the start time of the maximum instantaneous efficiency loss period, Indicates the end time of the maximum instantaneous efficiency loss period, when the maximum line instantaneous transport efficiency loss Reaching above 95% indicates entering a mid-term stable state.

4. The method for constructing a high-speed railway line risk map considering transportation efficiency loss according to claim 3, characterized in that: The risk event probability analysis includes station risk event probability analysis, interval risk event probability analysis and the spatiotemporal occurrence probability of the two types of events; The probability analysis of the station risk event is expressed as: in, represents the probability of a station risk event occurring on date type y, y=1 represents holidays, y=2 represents weekdays, y=3 represents weekends, N y (d w ) represents the number of late-starting events corresponding to the statistical date type y, N y (d c ) represents the number of station over-stop events corresponding to the statistical date type y, D y Indicates the number of days of the date type in the historical database, β indicates the date index, and n β Indicates the number of trains running on the line section within the date, path i，k represents a binary variable, 1 means that train number i passes through station k, and 0 means that train number i does not pass through station k. represents the scheduled departure time of train i at station k on date β, represents the planned arrival time of train i at station k on date β; When the train has a departure time at station k, the station path relationship is determined, and when the train has an arrival time at station k, the station path relationship is determined; The probability of the interval risk event is expressed as: in, represents the probability of occurrence of interval risk events under date type y, N y (d lt ) represents the number of temporary parking spaces in the corresponding statistical date type y, N y (d yw ) indicates the number of interval speed limit operation events under the corresponding statistical date type y, represents the binary decision variable of the interval (k, k+1) on the path of train number i, 1 means that train number i passes through the interval, 0 means that train number i does not pass through the interval, represents the scheduled departure time of train i at station k on date β, represents the planned arrival time of train i at station k+1 on date β; When the arrival time of the train at the interval station k+1 is greater than the departure time at the station k, it indicates that there is interval operation and the interval path relationship is determined; The spatiotemporal occurrence probabilities of the two types of events include the spatiotemporal occurrence probabilities of the departure delay event and the spatiotemporal occurrence probabilities of the interval temporary stop event; The spatiotemporal probability of the initial delay event is expressed as: ps∈[1,m] in, represents the spatiotemporal probability of a delayed event originating at time period pr and position ps under date type y, where pr represents the starting time of the time period, ps represents the position number of the initial disturbance of the risk, and P[E(d w )|E(d w ), E(d e )] represents the conditional probability of the initial delay event, Indicates the statistical percentage of departure delays in time period pr under date type y. represents the time distribution function of the initial delay event, (pr, pr+0.5) represents the integration period, It indicates the statistical proportion of delayed events originating at position ps under date type y. Indicates the frequency of the location of the originating delay event, represents the total number of departure delay events occurring at station k under date type y; The spatiotemporal probability of the temporary parking event in the section is expressed as: pS∈[1,m-1] in, represents the spatiotemporal probability of a temporary stop event in the starting interval at time period pr and location ps under date type y, P[E(d lt )|E(d lt ), E(d yw )] represents the conditional probability of temporary parking event in the starting section, It indicates the statistical proportion of temporary parking time in the starting section within the time period pr under date type y. represents the distribution function of temporary parking events in the originating section, It represents the statistical proportion of temporary parking events in the starting section at position ps under date type y. Indicates the frequency of temporary parking incidents at the originating section. Indicates the number of speed limit operation events that occurred in the interval (k, k+1) under date type y; When the position number ps takes the value [1, m-1], there are m-1 intervals in the corresponding line with m stations.

5. The method for constructing a high-speed railway line risk map considering transportation efficiency loss according to claim 4, characterized in that: The calculation of the potential risk impact of the space-time grid includes calculating the potential risk impact of the space-time grid on the station risk event and calculating the potential risk impact of the space-time grid on the interval risk event; The calculation of the potential risk impact of the space-time grid on the station risk event is expressed as: in, represents the quantitative value of the potential risk impact of the station in the spatiotemporal grid (pr, ps) under date type y, represents the average line transport efficiency loss caused by station risks in the spatiotemporal network (pr, ps), N y，z (pr, ps) represents the total number of station risk events in the space-time grid under date type y, represents the number of risk events at the rth level of the station in the spatiotemporal grid under date type y, lez represents the level of the risk event at the station, 10 4 represents the year-on-year expansion coefficient; The calculation of the potential risk impact of the space-time grid on the interval risk event is expressed as: in, It represents the quantitative value of the potential risk impact of the interval in the space-time grid (pr, ps) under the date type y. represents the average line transport efficiency loss caused by the risk within the interval of the space-time network (pr, ps), N y,q (pr, ps) represents the total number of risk events in the interval of the space-time grid under date type y, It represents the number of interval risk events of level r in the space-time grid under date type y, and leq represents the level of interval risk events.

6. The method for constructing a high-speed railway line risk map considering transportation efficiency loss according to claim 5, characterized in that: Drawing the route transport risk map includes reading data of the risk impact time-space distribution matrix, continuously interpolating the route transport risk, and drawing risk contour lines; The risk impact spatiotemporal distribution matrix is ​​expressed as: Among them, RF y Represents the spatial and temporal distribution matrix of the risk impact of the line under date type y.

7. The method for constructing a high-speed railway line risk map considering transportation efficiency loss according to claim 6, characterized in that: The risk influencing factor analysis of the risk map data includes sample attribute set construction and influencing factor analysis based on XGBOOST and LightGBM fusion; The sample attribute set construction includes the analysis of station risk event parameters and the analysis of interval risk event parameters; The station risk events include analyzing the quantitative value of station risk impact, the number of one-way arrival and departure lines, the station departure frequency and the station location; The section risk events include analyzing the quantified value of the section risk impact, the upper speed limit of the section, the density of train distribution in the section, the section location and the average speed grade coefficient of the vehicle type; The influencing factor analysis based on the fusion of XGBOOST and LightGBM includes importing risk factor analysis data, setting XGBOOST training model parameters, training based on XGBOOST, and loading the LightGBM analysis model.

8. A system for constructing a high-speed railway line risk map considering transportation efficiency loss based on any one of claims 1 to 7, characterized in that: include: Risk determination model, effectiveness quantification model, risk probability analysis model and risk mapping model; The risk determination model determines the types of risk events that occur during high-speed rail operation by identifying and classifying various risk events that affect train operation; The efficiency quantification model calculates the specific loss value of high-speed rail line transportation efficiency by quantifying the high-speed rail transportation efficiency loss caused by risk events; The risk probability analysis model performs a probability analysis of risk events based on the risk event type determination model and the transport efficiency loss quantification model, and calculates the frequency and probability of occurrence of different types of risk events within the statistical unit; The risk map drawing model draws a transportation risk map of the high-speed rail line based on the quantified value of the risk impact in the space-time grid, and conducts detailed data analysis based on the risk influencing factors to identify high-risk areas and time periods.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for constructing a high-speed railway line risk map considering transportation efficiency loss described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for constructing a high-speed railway line risk map considering transportation efficiency loss described in any one of claims 1 to 7 are implemented.

Images