CN111584091B

CN111584091B - Cross infection risk identification method and device for urban rail line-level close contact person

Info

Publication number: CN111584091B
Application number: CN202010357460.5A
Authority: CN
Inventors: 秦勇; 孙璇; 郭建媛; 谢臻; 高勃; 贾利民; 王雅观; 薛宏娇; 李健; 孙方
Original assignee: Beijing Jiaotong University
Current assignee: Beijing Jiaotong University
Priority date: 2020-04-29
Filing date: 2020-04-29
Publication date: 2023-10-24
Anticipated expiration: 2040-04-29
Also published as: CN111584091A

Abstract

The invention provides a method and a device for identifying cross infection risk of urban rail line-level close contact persons. Based on behavior science and data science, starting from microscopic travel behavior analysis, a risk exposure circuit station travel time probability distribution model is constructed; from the perspective of iteratable risk periods, the range of intimate contact persons with the possibility of cross-infection is continuously reduced; carrying out model solving by using an EM algorithm, estimating the risk probability of the cross-infection risk exposure train and the risk probability of the cross-infection of the close contact person, and establishing a risk grade evaluation model of the close contact person; and forming a urban rail risk level decision support and information query device. The invention not only is an important means for improving the screening precision and efficiency of suspected cases, but also provides a convenient and efficient query tool for the public to judge whether the public is an 'intimate contact', fills the gap of risk identification of the intimate contact in the public transportation field, and has important decision support significance for the public emotion guiding and emergency resource deployment under emergency.

Description

Cross infection risk identification method and device for urban rail line-level close contact person

Technical Field

The invention relates to the technical field of rail transit passenger health management, in particular to a method and a device for identifying cross infection risk of urban rail line-level close contact persons.

Background

Urban rail transit (hereinafter referred to as urban rail) is an indispensable tool for public travel as a backbone strength for public transportation operation inside urban areas. The major sudden public health event has the characteristic of close contact-high propagation, and once the major sudden public health event occurs under the urban rail networking operation condition, the risk propagation speed is high, the propagation range is wide, and uncontrollable risks are easy to generate. However, public transportation means such as urban rails, airplanes, railways, buses and the like have an accurate identity recognition function of one ticket/machine, and if a definite diagnosis or suspected case (hereinafter referred to as a case) goes on the urban rails, a plurality of trains can be checked in a time range of entering and exiting, and a closely contacted person (hereinafter referred to as a closely contacted person) of the case cannot be quickly screened under the condition that the checked-in trains are not known only according to a physical path. Therefore, in the period of limited prevention and control measures before the case is not isolated, if the closely connected person can be rapidly identified and the cross infection risk level can be confirmed, the method has great decision support significance for providing accurate research and judgment, joint prevention and control, public emotion guiding and emergency resource deployment for related departments.

At present, the prior art only relates to the track tracking and risk type division of the closely-connected people in public transportation means such as airplanes, railways, buses and the like aiming at epidemic situation isolation management and track tracking related methods of the closely-connected people, only relates to basic information collection and release stages aiming at public transportation means such as urban rails and the like, and no method or tool for automatically identifying the cross infection risk of urban rail line-level closely-connected people based on data driving is found. Moreover, the method in the prior art is only suitable for cases traveling on only one line of the urban rail network, does not relate to transfer and a plurality of lines, does not relate to risk identification of transfer passenger flows in the lines, and cannot screen close-connected persons with the urban rail network level as a range.

Disclosure of Invention

The embodiment of the invention provides a method and a device for acquiring cross infection risk level of a rail transit intimate contact person, which are used for solving the problems of the prior art.

In order to achieve the above purpose, the present invention adopts the following technical scheme.

According to one aspect of the present invention, there is provided a method for identifying risk of cross infection of urban rail line-level intimate contact, comprising:

step S1, acquiring a risk exposure line REL, a risk exposure physical path RER and travel duration of a case;

Step S2, analyzing the microscopic travel behaviors and travel time of the case according to REL, RER and travel time length of the case, and acquiring the travel physical path, microscopic travel behaviors and travel process of the close contact person;

step S3, determining a RESFC set, a RETFC set and a REPFC period of a cross-infection risk exposure station according to microscopic travel behaviors and travel processes of the case and travel physical paths, microscopic travel behaviors and travel processes of the packer; the close connector is a urban rail line-level close contact connector;

s4, constructing a cross infection risk exposure physical path RERFC set and an effective train ET set of the packer, giving a unique identification code to one piece of data, and constructing a packer initial data set Dataset (0) containing the unique identification code of the case;

s5, constructing a passenger travel time probability distribution model of each station in a risk exposure line REL containing hidden variables based on a data set Dataset (0) of the initial data set of the close connector;

step S6, solving a passenger travel time probability distribution model based on an expectation maximization EM algorithm, and calculating distribution parameters of the passenger group' S journey-ascending train probability and the passenger outbound travel time;

Step S7, calculating the distribution parameters of the passenger arrival and departure time of each station according to the probability of the passing trains of the passenger group and the distribution parameters of the passenger arrival and departure time, and calculating the probability of cross infection risk of the trains in the RETFC set and the probability of passing trains of the close connector;

step S8, updating the cross infection risk exposure period REPFC in the step 3 based on the cross infection risk probability of the trains in the RETFC set and the joint connector train journey probability, and updating the Dataset (0) into a cross-infection risk data set Dataset (new) based on the updated REPFC, wherein the unique identification code is unchanged;

step S9, calculating the cross infection risk probability POC of the close-connected person based on a cross infection risk data set Dataset (new);

and S10, determining the cross infection risk level of the close-contact person according to the cross infection risk probability POC of the close-contact person and a set cross infection risk level classification strategy.

Preferably, the risk exposure line REL, the risk exposure physical path RER, and the trip duration of the acquired case in the step S1 specifically include:

step S011, determining the intelligent traffic card number, travel time period and travel origin-destination (OD) pair information of the case;

step S012, inquiring basic information of a line station in a database, and determining the trip OD pair codes of the case as follows:

The risk exposure physical path RER of the case is:

step S013, inquiring the AFC data of the passengers in the database by taking the intelligent traffic card number and the travel OD pair of the case as indexes, and determining the case record and the corresponding travel time wherein ,/>Determining the travel duration of the case according to the travel time>

Preferably, in the step S2, microscopic trip behaviors and trip time of the case are analyzed according to REL, RER and trip time of the case, and the method specifically includes:

step S021, analyzing microscopic trip behaviors contained in trip links of cases, wherein the microscopic trip behaviors comprise: the trip OD pair selection behavior, the inbound card swiping behavior, the inbound travel behavior, the inbound waiting behavior, the inbound train selection behavior, the outbound travel behavior and the outbound card swiping behavior;

step S022, obtaining travel time T corresponding to the microscopic travel behavior of the case, including: inbound travel time AT, inbound waiting time WT, riding time BT, and outbound travel time ET, i.e., t=at+wt+bt+et;

preferably, in the step S3, the determining the ESFC set, the refc set and the refc according to the microscopic trip behavior and trip process of the case and the trip physical path, the microscopic trip behavior and the trip process of the contact specifically includes:

Step S031, according to the basic information of the line site, inquiring REL upper homodromous site after the travel starting point of the case, and constructing the RESFC set

wherein ,indicating the start of a case on a risk exposure line, l _i Representing the ith site on line l;

step S032, encoding according to the case ODTravel time-> wherein ,/>The train is in +.>The RETFC set is constructed according to the arrival time and the departure time of the mobile terminal:

wherein ,tⁱ Is the i-th cross-infection risk exposure train, and all trains in the RETFC set meet the following constraint conditions:

(1) Minimum inbound travel time constraint of the start station, and the train is larger than the inbound card swiping time at the departure time of the start station

wherein ,representing train t ⁱ At case p origin o ^p Is to go round>Indicating that case p is at origin o ^p Is->Represents o ^p Minimum inbound travel time of a station, t represents the maximum number of cross-infection risk exposure trains;

(2) And the minimum outbound running time constraint of the terminal. The time of the train arriving at the terminal station is smaller than the time of the card swiping at the outlet station:

wherein ,indicating that case p is at destination d ^p Is->Representing train t ⁱ At case p terminal d ^p Arrival time of- >Representation d ^p Minimum outbound travel time of a station, t representing the maximum number of cross-infection risk exposure trains;

step S033, determining REPFC, T according to each train AVL data in the RETFC set ^repfc ＝[T ^start ,T ^end ]The following conditions are satisfied:

wherein ,representing train t ¹ Origin station l on REL in the same direction as case p ₁ Arrival time,/->Indicating the originating station l ₁ Maximum inbound travel time of (2); />Representing train t ^t Terminal l on REL in the same direction as case p _n Departure time, ->Indicating terminal station l _n Is a maximum inbound travel time of (c).

Preferably, the step S4 constructs a set of cross-infection risk exposure physical paths refcs and a set of effective trains ET of the packer, and specifically includes:

step S041, determining RERFC set

Step S042, inquiring AFC data of an automatic fare collection system of urban rail transit in a database, determining an initial Dataset Dataset (0) of a packer, including AFC data of a case, giving each record a unique identification code, wherein each record comprises an identification code, an ID card number, an inbound card swiping time, an outbound card swiping time, an inbound station number and an outbound station number;

and step S043, imitating the RETFC set constructing method in the step S032, and constructing the set of the closely connected persons ET according to the code of the pair of passengers OD and the traveling time and combining the arrival time and the departure time of the train at the O and D sites in the automatic train positioning data.

Preferably, the constructing a passenger travel time probability distribution model of each station in the risk exposure line REL containing hidden variables based on the initial Dataset (0) of the packer in the step S5 specifically includes:

step S051, assuming station travel time distribution, setting the station travel time of all the passengers on the REL to be independent probability density distribution, wherein the passenger group comprises cases and close-connected persons;

step S052, passenger classification assumption of arrival according to T ^repfc ＝[T ^start ,T ^end ]Total period and RERFC setThe data set database (0) of the initial data set of the packer is divided into N classes according to the time granularity delta t, wherein the N classes comprise:

(1) Conversion of the arrival card swiping time of the same OD to the jth individual observation data in the kth passenger groupAnd the outbound card swiping time->Time format expressed in seconds

(2) The packer initial Dataset (0) is partitioned into N classes,

N＝T ^end -T ^start /Δt (9)

step S053, unknown hidden variables are defined according to the passing trains of passengers in the platform, and the hidden variables comprise:

(1)UD ⁱ unobserved data representing class k passenger population

(2) By UD _j ^k Unobserved data representing the jth individual in a kth passenger population

wherein ,

(3)π ^k represents the probability of a passing train for a class k passenger group,representing probability of boarding a t train for a k-th passenger group

(4) Supplementing the packer's initial dataset with complete dataSet satisfactionThe method comprises the following steps:

step S054, representing parameters to be estimated according to the probability density distribution parameters in step S051 and the probability variables of the passenger boarding train in step S053, wherein the parameters to be estimated comprise:

(1) Constructing parameters to be estimated for a kth passenger population

wherein ,representing the probability of boarding the ith train for the kth passenger group, < >>Representing site l _s Internal outbound travel time parameters;

(2) Construction model parameter θ= (θ) ¹ ,…,θ ^k ,…,θ ^N ) ^(step) ；

Step S055, constructing likelihood function of passenger travel time probability distribution model containing hidden variable

Step S056, converting likelihood function into log likelihood function

Step S057, maximizing log likelihood function

L(θ)＝maxL _c (θ)＝max(logP(t ⁱⁿ ,t ^out ,UD|θ)) (20)

Preferably, in the step S6, the passenger travel time probability distribution model is solved based on the expectation maximization EM algorithm, and distribution parameters of the passenger group' S on-journey train probability and the passenger outbound travel time are calculated, which specifically includes:

step S061, determining the initial value θ=θ of the parameter ⁽⁰⁾ ；

Step S062, E-step calculation is performed, and the observation data (t) ⁱⁿ ,t ^out ) At θ=θ ^(step) Lower calculation Q function Q (θ)

Q(θ)＝E(L _c (θ)|(t ⁱⁿ ,t ^out ),θ ^(step) ) (21)

wherein ,θ^(step) Is the estimated value of step theta of the algorithm;

step S063, performing M-step calculation to maximize Q function Q ^(step) (θ) updating the parameter estimate θ ^(step+1) ＝arg maxQ ^(step) (θ)

Wherein, the probability of the passing train of the kth passenger group in step +1 is thatCalculating distribution parameters of passenger arrival travel time by using optimization method of L-BFGS-B>

Step S064, step S062 and said step S063 are performed alternately until the algorithm converges.

Preferably, in the step S7, the distribution parameters of the passenger arrival/departure time of each station, the cross infection risk probability of the trains in the RETFC set, and the joint person train arrival probability are calculated according to the distribution parameters of the passenger arrival/departure time and the passenger departure time, and specifically include:

step S071 of obtaining final parameter estimation resultAnd outbound running time distribution parameters of stations of risk line REL

wherein ,representing the probability of a passing train of a kth passenger group; />Representing site l _s The passenger outbound time distribution parameters;

step S072, calculating each site l of the risk exposure line according to the AFC data among the OD pairs _s Passenger outbound travel time distribution parameters of (c)

Step S073, determining each station l of the risk exposure line according to the outbound running time distribution of the passengers and the inbound running time distribution curve of the station _s Passenger arrival travel time parameter of (2)

And distinguishing the case from the packer according to the unique identification codes, and identifying the cross infection risk probability and the packer train routing probability of the trains in the RETFC set.

Preferably, the updating the cross-infection risk exposure period reffc in step S8 updates the database (0) to the cross-infected person risk data set database (new), and the unique identification code is unchanged, which specifically includes:

exposing each station l of the line according to the risk _s The passenger arrival and departure time distribution parameter and the calculation formula T of REPFC ^repfc ＝[T ^start ,T ^end ]Updating the REPFC, shortening the minimum inbound card swiping time and the maximum outbound card swiping time of Dataset (0) based on the updated time length of the REPFC, and updating a cross-infected person risk data set Dataset (new), wherein the total time length of the Dataset (new) is smaller than Dataset (0).

Preferably, the calculating the risk probability of cross infection POC of the close contact person based on the risk Dataset of cross-infection person (new) in step S9 specifically includes:

step S091, extracting case data according to the unique identification code, and confirming risk probability of each train in RETFC set

wherein ,indicating that case p belongs to the kth passenger group, the trains in the range RETFC set are +.>Risk probability values for (a);

step S092, collecting train numbers according to the RETFC All ascending trains +.>Extracting +.about.of the ascending train>Probability of risk

wherein ,indicating that the close contact person j belongs to the kth passenger group, and the train in the range RETFC set is +.>Probability values of (2);

step S093, calculating the probability of risk of cross infection POC of the packer

According to another aspect of the present invention, there is provided a risk identification device for cross infection of urban rail line-level intimate contact, comprising:

the information acquisition module is used for acquiring a risk exposure line REL, a risk exposure physical path RER and a travel time length of the case, analyzing the microscopic travel behaviors and travel time of the case according to the REL, RER and travel time length of the case, and acquiring the travel physical path, microscopic travel behaviors and travel process of the close contact person;

the data analysis and processing module is used for determining a cross infection risk exposure site RESFC set, a cross infection risk exposure train RETFC set and a cross infection risk exposure period REPFC according to the microscopic travel behaviors and travel processes of the cases and the travel physical paths, the microscopic travel behaviors and the travel processes of the close-connected people; constructing a cross infection risk exposure physical path RERFC set and an effective train ET set of the close connector, giving a unique identification code to one piece of data, and constructing an initial Dataset (0) of the close connector containing the unique identification code of the case;

The model construction and calculation module is used for constructing a passenger travel time probability distribution model of each station in the risk exposure line REL containing hidden variables based on the initial Dataset Dataset (0) of the packer; solving a passenger travel time probability distribution model based on an Expectation Maximization (EM) algorithm, and calculating the distribution parameters of the journey-ascending train probability of the passenger group and the passenger outbound travel time; calculating the distribution parameters of the passenger arrival and departure time of each station according to the probability of the passing trains of the passenger group and the distribution parameters of the passenger arrival and departure time, and calculating the probability of cross infection risk of the trains in the RETFC set and the probability of passing the trains of the close connector;

updating the cross infection risk exposure period REPFC in the step 3 based on the cross infection risk probability of the trains in the RETFC set and the joint connector train journey probability, and updating the database (0) into a cross infection risk data set (new) based on the updated REPFC, wherein the unique identification code is unchanged; the risk of cross-infection probability POC of the fitter is calculated based on the risk Dataset of cross-infectors Dataset (new).

The decision support module is used for setting the risk level of cross infection and issuing information;

and the risk level query module is used for determining the cross infection risk level of the close-connected person according to the POC and the set cross infection risk level classification strategy and providing a public query function.

According to the technical scheme provided by the embodiment of the invention, the cross infection risk level acquisition method of the rail transit line-level intimate contact person is based on statistics, behavior science and data science, and starts from microscopic trip behavior analysis, a risk train set generation method and a log-in train matching generation method are provided, and a risk exposure line station travel time probability distribution model containing hidden variables is constructed; from the angle of the iterative risk period window, the screening range of the close contact person with the possibility of cross infection is continuously narrowed; and carrying out model solving by using an Expectation Maximization (EM) algorithm, estimating the risk probability of the cross-infection risk exposure train and the cross-infection risk probability of the close contact person, and establishing a risk grade assessment model of the close contact person on the basis.

The method provided by the embodiment of the invention can be suitable for the scene that the case goes out on a plurality of lines and is transferred on different lines, and can be used for screening the close-connected persons in a range of urban railway network level to obtain the cross infection risk level of the transfer passenger flow in different lines.

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

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a process flow diagram of a method for acquiring a cross infection risk level of a rail transit intimate contact person according to an embodiment of the present invention.

Fig. 2 is a cross-infection risk line diagram of a case provided in an embodiment of the present invention.

Fig. 3 is a schematic diagram of microscopic behavior analysis of a subway-climbing train scheme according to an embodiment of the present invention.

Fig. 4 is a graph showing the running time profile of the outbound site for a confirmed or suspected case according to an embodiment of the present invention.

FIG. 5 is a graph of risk probability levels for an intimate contact according to an embodiment of the present invention;

fig. 6 is a specific structural diagram of a cross infection risk identification device for urban rail line-level intimate contact according to an embodiment of the present invention.

Detailed Description

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for explaining the present invention and are not to be construed as limiting the present invention.

As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

For the purpose of facilitating an understanding of the embodiments of the invention, reference will now be made to the drawings of several specific embodiments illustrated in the drawings and in no way should be taken to limit the embodiments of the invention.

Example 1

The embodiment of the invention provides a risk identification method of a data-driven urban rail transit line-level intimate contact (hereinafter referred to as a packer), wherein the implementation principle schematic diagram of the method is shown in fig. 1, and the method comprises the following processing steps:

step S1, acquiring risk exposure lines (Risk exposure Line, REL), risk exposure physical paths (Risk exposure route, RER) and travel time length of cases;

the risk exposure line REL is a case riding line, the risk exposure physical path RER is a section through which the case travels on the OD pair, for example, the case travels on the fourth line, the OD pair is the siemens-safe, the risk exposure line REL is the 4 line, and the risk exposure physical path RER is the section of the siemens-safe.

S2, analyzing microscopic travel behaviors and travel time of cases, and relating to travel physical paths of the close-connected persons, microscopic travel behaviors and travel process analysis method cases;

step S3, determining a set of cross-infection risk exposure sites (Risk exposure station for cross-in, RESFC), a set of cross-infection risk exposure trains (Risk exposure train for cross-in, RETFCs) and a cross-infection risk exposure period (Risk exposure period for cross-in, REPFC);

Step S4, constructing a set of cross infection risk exposure physical paths (RERFCs) of the close-connected persons, and an Effective Train (ET) set, wherein the Effective train ET set constructing mode is the same as the RETFC set constructing mode in the step 3, and constructing an initial data set (0) of the close-connected persons, wherein the initial data set contains unique identification codes of cases;

and S5, constructing a passenger travel time probability distribution model of each station in the risk exposure line REL containing hidden variables based on the initial data set Dataset (0) of the close connector.

And S6, solving a passenger travel time probability distribution model based on an Expectation-maximization (EM) algorithm, and calculating distribution parameters of the passenger group' S journey train probability and the passenger outbound travel time.

Step S9, calculating the risk probability of cross infection (Probability of cross infection, POC) of the close-coupled person based on the risk Dataset of cross-infected person (new);

In the specific embodiment of the present invention, step 1 obtains a risk exposure line REL, a risk exposure physical path RER, and a trip duration of a case, and specifically includes:

step S011, determining the intelligent traffic card number, travel time period and travel Origin (OD) peer-to-peer information of the case according to the forms of case reporting, network gathering, newspaper consulting and telephone confirmation of publications and the like;

step S012, inquiring basic information of a line station in a rail transit line basic information database, and determining that the travel OD pair of the case is encoded as follows:

the risk exposure physical path RER of the case is:

wherein ,indicated as patient p outbound on line l +.>Arrival stationIs->Is a travel path of the vehicle; l (L) _m Representing all site sets on line l; l (L) _o 、l _d Indicating a certain station on line l.

Step S013, inquiring the data of the transaction records (Automatic Fare Collection, AFC) of the passenger card swiping in the database by taking the intelligent traffic card number and the travel OD pair of the case as indexes, and determining the case records and the corresponding travel time ( wherein ,/>) Determining the travel duration of the case according to the travel time>

In the specific embodiment of the invention, the step 2 of analyzing the microscopic trip behaviors and trip time of the case specifically comprises the following steps:

step S021, analyzing microscopic trip behaviors contained in trip links of cases, wherein the step S021 comprises the following steps: the trip OD pair selection behavior, the inbound card swiping behavior, the inbound travel behavior, the inbound waiting behavior, the inbound train selection behavior, the outbound travel behavior and the outbound card swiping behavior;

step S022, the micro travel behaviors between the OD pairs of the cases in step 021 correspond to four travel times, wherein the steps comprise: an Access Time (AT), an Access Waiting Time (WT), a riding Time (BT), and an Exit Time (ET), i.e., t=at+wt+bt+et;

in a specific embodiment of the present invention, step 3 of identifying a set of cross-infection risk exposure sites RESFC, a set of cross-infection risk exposure trains refc, and a cross-infection risk exposure period refc specifically includes:

step S032, encoding according to the case ODTravel time->( wherein ,/>) Train in combination with automatic train positioning (Automatic Vehicle Location, AVL) data>Is to construct the RETFC set according to the arrival time (Train Arrive Time, TAT) and the departure time (Train Departure Time, TDT)

wherein ,indicating that case p is at destination d ^p Is->Representing train t ⁱ At case p terminal d ^p Arrival time of->Representation d ^p Minimum outbound travel time of a station, t representing the maximum number of cross-infection risk exposure trains;

Step S033, determining REPFC, T according to the automatic train positioning data in the RETFC set ^repfc ＝[T ^start ,T ^end ]The following conditions are satisfied:

wherein ,representing train t ¹ Origin station l on REL in the same direction as case p ₁ Arrival time,/->Indicating the originating station l ₁ Maximum inbound travel time of (2); />Representing train t ^t Terminal l on REL in the same direction as case p _n Departure time, ->Indicating terminal station l _n Maximum inbound travel time of (2);

in the specific embodiment of the invention, the step 4 is to construct a set of cross infection risk exposure physical paths RERFCs and a set of effective trains ET of the close-connected people, and specifically comprises the following steps:

step S041, determining RERFC set

Step S042, inquiring AFC (Automatic Fare Collection System, referring to an automatic fare collection system for urban rail transit) data in a database, determining an initial Dataset Dataset (0) of a packer, including AFC data of a case, and giving each record a unique identification code, wherein each record comprises an identification code, an ID card number, an inbound card swiping time, an outbound card swiping time, an inbound station number and an outbound station number;

step S043, imitating the RETFC set constructing method involved in the step S032, and constructing the set of the closely connected persons ET according to the code of the pair of passengers OD and the traveling time and combining the arrival time and the departure time of the train at the O site and the D site in the automatic train positioning data;

In a specific embodiment of the present invention, the constructing a passenger travel time probability distribution model of each station in the risk exposure line REL containing the hidden variable based on the initial Dataset (0) in step 5 specifically includes:

step S051, the station travel time distribution assumption, namely the case and the packer are regarded as passengers, the passenger is used for representing the case, the arrival and departure travel time of all the station passengers on REL are independent and distributed, and the assumption can be verified to be a certain probability density distribution, such as normal distribution, gamma distribution and the like, according to experience or actual data;

step S052, passenger classification assumption of arrival according to T ^repfc ＝[T ^start ,T ^end ]Total period and RERFC setThe OD pairs of (a) and (b) are divided into N classes according to the time granularity Δt, wherein the N classes comprise:

(2) The packer initial Dataset (0) is partitioned into N classes,

N＝T ^end -T ^start /Δt (9)

step S053, defining hidden variables according to unknown boarding events of passengers in the platform, wherein the hidden variables comprise:

(1)UD ⁱ unobserved data representing class k passenger population

(2) By usingUnobserved data representing the jth individual in a kth passenger population

wherein ,

/>

(3)π ^k representing the probability of a journey for a class k passenger group,representing probability of boarding a t train for a k-th passenger group

(4) Supplementing the initial dataset of the packer with the full dataset satisfiesThe method comprises the following steps:

(1) Constructing parameters to be estimated for a kth passenger population

Step S056, converting likelihood function into log likelihood function

Step S057, maximizing log likelihood function

L(θ)＝maxL _c (θ)＝max(logP(t ⁱⁿ ,t ^out ,UD|θ)) (20)

In a specific embodiment of the present invention, step 6 solves a passenger travel time probability distribution model based on an expectation maximization EM algorithm, and calculates distribution parameters of a passenger group's on-haul train probability and passenger outbound travel time, and specifically includes:

step S061, determining the initial value θ=θ of the parameter ⁽⁰⁾ ；

Q(θ)＝E(L _c (θ)|(t ⁱⁿ ,t ^out ),θ ^(step) ) (21)

wherein ,θ^(step) Is the estimated value of step theta of the algorithm;

Step S064, step S062 and said step S063 are alternately carried out until the algorithm converges;

in the specific embodiment of the present invention, in step 7, the distribution parameters of the passenger arrival and departure times of each station, the risk probability of cross infection of the trains in the RETFC set, and the probability of the contact person train arrival are calculated according to the distribution parameters of the passenger arrival and departure times of the passenger group, and specifically include:

Step S073, determining the passenger 'S arrival time parameters according to the passenger' S arrival time distribution fitting station arrival time distribution curve

In a specific embodiment of the present invention, the updating the cross infection risk exposure period (Risk exposure period for cross-input, reffc) in step 8, updating the database (0) to be the cross-infected person risk data set database (new), the unique identification code is unchanged, specifically including:

updating REPFC according to the parameters obtained in the steps S072 and S073 in the step 7 and the REPFC calculation formula described in the step S033, shortening the minimum inbound card swiping time and the maximum outbound card swiping time of Dataset (0) based on the updated time length of REPFC, and updating a cross-infection person risk data set Dataset (new), wherein the total length of the obtained Dataset (new) time is smaller than Dataset (0);

in a specific embodiment of the present invention, calculating the risk probability of cross-infection POC of the fitter based on the risk Dataset of cross-infectors (new) in step 9 specifically includes:

/>

wherein ,indicating that case p belongs to the kth passenger group, the trains in the range RETFC set are +. >Risk probability values for (a);

step S092, collecting train numbers according to the RETFCAll ascending trains +.>Extracting +.about.of the ascending train>Probability of risk

wherein ,indicating that the packer j belongs to the kth passenger group, and a set of RETFCs is loggedMiddle train->Probability values of (2);

step S093, calculating the risk probability of cross infection of the packer

In a specific embodiment of the present invention, determining the risk level of cross infection of the packer according to the probability of risk of cross infection POC of the packer in step 10 specifically includes:

dividing four risk classes of low cross infection risk, general cross infection risk, high cross infection risk and extremely high cross infection risk, wherein the risk probability value ranges corresponding to the low cross infection risk, general cross infection risk, high cross infection risk and extremely high cross infection risk respectively are as follows: { (0,0.25 ], (0.25, 0.5], (0.5, 0.75], (0.75,1);

the cross infection risk level of the close-connected person is determined according to the risk probability value range corresponding to the value of the cross infection risk probability POC of the close-connected person: low risk of cross infection, general risk of cross infection, high risk of cross infection, or extremely high risk of cross infection.

Example two

The risk level of an intimate contact person (hereinafter referred to as a packer) is automatically identified by taking a subway exit behavior example of a definite diagnosis or suspected case (hereinafter referred to as a case) in Beijing city, and the process of the invention is as follows:

1. obtaining a case risk exposure line REL as a certain line [ l ] according to a case card number 5XXXXXX9 ₆₉ ,l ₆₁ ]The risk exposure physical path RER isThe arrival time is 08:17:00 of a month of a year, the conversion seconds are 29820, the arrival time is 08:31:03 of the same day, the conversion seconds are 30663, the travel time length is 14min, and the cross infection risk circuit diagram RER is shown in fig. 2;

2. the microscopic trip behavior and trip process composition of the case are analyzed, and the starting station is the first station l ₆₉ The direction is downstream, RER is a line-level transfer-free path, and the line-level transfer-free path totally comprises 10 stations { l ₆₉ ,l ₆₅ ,l ₆₃ ,l ₆₁ ,l ₅₉ ,l ₅₇ ,l ₅₅ ,l ₅₃ ,l ₅₁ ,l ₄₉ -a }; the microscopic trip behaviors include: the trip OD pair selection behavior, the inbound card swiping behavior, the inbound travel behavior, the inbound waiting behavior, the inbound train selection behavior, the outbound travel behavior and the outbound card swiping behavior; the corresponding travel time includes: the arrival travel time AT, the arrival waiting time WT, the riding time BT and the departure travel time ET are also suitable for the travel analysis of the close-up person, the travel time composition is described by taking the first vehicle of case energy to be logged on as an example, and a microscopic behavior analysis schematic diagram of a subway logging train scheme is shown in fig. 3;

3. Determining a cross-infection risk exposure site RESFC setSince the case start station is the start station, the stations after the start station are all listed as RESFC aggregate elements with the destination as the destination;

determining a RETFC set of the cross-infection risk exposure train as according to the AFC data and the AVL dataNamely, the case has two trains which can be logged in the subway trip at the time;

setting initial maximum outbound-inbound travel time, and determining a cross-infection risk exposure period REPFC to be T ^repfc(0) ＝[07:48:00，9:20:50]I.e. [28080s,33650s ]]；

4. Determining a physical path RERFC set exposed by cross infection risk of subways of the close connector, wherein the case starting station is the originating station, so that the OD pairs in the downlink direction formed by all stations on the line are included, and the total number is 10 x 9/2=45 pairs;

5. constructing a packer initial data set Dataset (0) containing a unique identification code of a case, wherein the packer initial data set contains 7744 pieces of AFC data in total of case AFC data;

6. constructing a passenger travel time probability distribution model of each station of the risk exposure line containing hidden variables, and assuming that passenger travel time distribution obeys gamma distribution; the accuracy of the card swiping time of the passenger in the station is 1min, and the same OD is divided for the passenger for the window in 1min, namely, Δt=1min=60 s;

7. based on the EM algorithm solving model, the parameter estimation result of the case is that The distribution of the running time of the outbound site is shown in figure 4, and the parameter estimation result values of other individuals are +.>Not specifically enumerated herein;

8. risk probability for each train in RETFC setAccording to the set of contact ET +.>Any train number can be used for determining that the total number of the close-connected people on the same line is 788;

9. updating REPFC to T according to the time distribution probability density function of the incoming and outgoing station running time of the line starting station and the line terminating station and the REPFC calculation formula ^repfc(new) ＝[07:56:00，08:56:50]I.e. [28560s,32210s ]]Updating Dataset (0) into a cross-infection person risk data set Dataset (new), wherein the total number of the Dataset (0) is 5069 AFC data, and the total number of individuals with the possibility of cross-infection risk is reduced to 767;

10. estimating a risk probability value according to a joint packer cross infection risk probability formula, and determining 635 persons in the same line joint packer according to the risk probability value;

11. the risk level of cross infection of the close-connected person is determined according to the POC and the set risk level classification strategy of cross infection, and the risk level distribution situation of the close-connected person provided by the embodiment of the invention is shown in figure 5.

Example III

This embodiment provides a cross infection risk identification device for urban rail line-level intimate contact, which has a specific structure as shown in fig. 6 and comprises:

The information acquisition module 61 is subordinate to the back-end processing module and is used for acquiring a risk exposure line REL, a risk exposure physical path RER and a travel duration of a case, analyzing the microscopic travel behavior and travel time of the case according to the REL, the RER and the travel duration of the case, and acquiring the travel physical path, the microscopic travel behavior and the travel process of the close contact person;

the data analysis and processing module 62 is subordinate to the back-end processing module and is used for inquiring the number of the case risk exposure line REL, the number of the risk exposure physical path RER, the number of the cross-infection risk exposure site RESFC, the number of the cross-infection risk exposure train refc train number and the number of the cross-infection risk exposure physical path RERFC set OD pair number of the packer; for preprocessing the AFC data of the packer data set. Determining a RESFC set, a RETFC set and a REPFC period according to microscopic trip behaviors and trip processes of the cases and trip physical paths, microscopic trip behaviors and trip processes of the close-connected people; constructing a cross infection risk exposure physical path RERFC set and an effective train ET set of the close connector, giving a unique identification code to one piece of data, and constructing an initial Dataset (0) of the close connector containing the unique identification code of the case;

The model building and calculating module 63 is subordinate to the back-end processing module and is used for building a passenger travel time probability distribution model of each station in the risk exposure line REL containing hidden variables based on the initial data set Dataset (0) of the packer; solving a passenger travel time probability distribution model based on an Expectation Maximization (EM) algorithm, and calculating the distribution parameters of the journey-ascending train probability of the passenger group and the passenger outbound travel time; calculating the distribution parameters of the passenger arrival and departure time of each station according to the probability of the passing trains of the passenger group and the distribution parameters of the passenger arrival and departure time, and calculating the probability of cross infection risk of the trains in the RETFC set and the probability of passing the trains of the close connector;

The decision support module 64 is subordinate to the back-end processing module and is used for setting the risk level of cross infection and carrying out decision support such as information release, emergency command, public emotion dispersion and the like;

The risk level query module 65 is subordinate to the front-end information issuing module and is used for assisting the public to query whether the public is a close-connected person. And determining the cross infection risk level of the close-connected person according to the cross infection risk probability POC of the close-connected person and a set cross infection risk level classification strategy.

Preferably, the specific processing procedure of the information obtaining module 61 to obtain the risk exposure line REL, the risk exposure physical path RER and the trip duration of the case includes:

the risk exposure physical path RER of the case is:

Preferably, the information obtaining module 61 analyzes the microscopic trip behavior and trip time of the case according to the REL, RER and trip time of the case, and the specific processing procedure includes:

preferably, the data analysis and processing module 62 determines the ESFC set, the refc set and the refc according to the microscopic trip behavior and trip process composition of the case and the trip physical path, the microscopic trip behavior and trip process of the packer, and the specific processing process includes:

wherein ,indicating the start of a case on a risk exposure line, l _i Representing the first line on line l _i A site;

Preferably, the data analysis and processing module 62 constructs a set of cross-infection risk exposure physical paths RERFCs and a set of effective trains ET of the close-connected people, and the specific processing process includes:

Step S041, determining RERFC set

Preferably, the model building and calculating module 63 includes a model building module and a model calculating module.

The model building module is configured to build a passenger travel time probability distribution model of each station in the risk exposure line REL containing hidden variables based on the initial Dataset (0) of the packer, and the specific processing procedure includes:

Step S052. The passenger classification hypothesis of coming in according to T ^repfc ＝[T ^start ,T ^end ]Total period and RERFC setThe data set database (0) of the initial data set of the packer is divided into N classes according to the time granularity delta t, wherein the N classes comprise:

(2) The packer initial Dataset (0) is partitioned into N classes,

N＝T ^end -T ^start /Δt (8)

(1)UD ⁱ unobserved data representing class k passenger population

wherein ,

UD＝(UD ¹ ,…,UD ^k ,…,UD ^N ) (15)

(1) Constructing parameters to be estimated for a kth passenger population

Step S056, converting likelihood function into log likelihood function

Step S057, maximizing log likelihood function

L(θ)＝maxL _c (θ)＝max(logP(t ⁱⁿ ,t ^out ,UD|θ)) (19)

Preferably, the model operation module solves a passenger travel time probability distribution model based on an expectation maximization EM algorithm, calculates distribution parameters of the passenger group's journey-ascending train probability and the passenger outbound travel time, and the specific processing procedure includes:

step S061, determining the initial value θ=θ of the parameter ⁽⁰⁾ ；

Q(θ)＝E(L _c (θ)|(t ⁱⁿ ,t ^out ),θ ^(step) ) (20)

wherein ,θ^(step) Is the estimated value of step theta of the algorithm;

Wherein, the probability of the passing train of the kth passenger group in step +1 is thatCalculating distribution parameters of passenger arrival travel time by using optimization method of L-BFGS-B>/>

Preferably, the model operation module calculates the distribution parameters of the passenger arrival and departure time of each station according to the distribution parameters of the arrival and departure train probability and the passenger departure time of the passenger group, and the cross infection risk probability and the packer train arrival probability of the trains in the RETFC set, and the specific processing process comprises:

step S072, calculating a risk exposure line according to the AFC data among the OD pairsRoad each station l _s Passenger outbound travel time distribution parameters of (c)

Preferably, the model operation module updates the cross infection risk exposure period reffc, updates the database (0) to a cross-infected person risk data set database (new), and the unique identification code is unchanged, and the specific processing procedure includes:

Preferably, the model operation module calculates the risk probability of cross infection POC of the close-coupled person based on a risk Dataset of cross-infected person (new), and the specific processing procedure includes:

Preferably, the risk level query module 65 determines the risk level of cross infection of the packer according to the probability of risk of cross infection POC of the packer, and specifically includes:

The specific process of performing cross infection risk level of the rail transit intimate contact person by using the device of the embodiment of the present invention is similar to the previous method embodiment, and will not be repeated here.

In summary, the method for acquiring the cross infection risk level of the rail transit line-level intimate contact according to the embodiment of the invention starts from microscopic trip behavior analysis based on statistics, behavior science and data science, provides a risk train set generation method and a log-on train matching generation method, and constructs a running time probability distribution model of each station of a risk exposure line containing hidden variables; from the angle of the iterative risk period window, the screening range of the close contact person with the possibility of cross infection is continuously narrowed; and carrying out model solving by using an Expectation Maximization (EM) algorithm, estimating the risk probability of the cross-infection risk exposure train and the cross-infection risk probability of the close contact person, and establishing a risk grade assessment model of the close contact person on the basis.

The method provided by the embodiment of the invention is an important means for improving the screening precision and efficiency of suspected cases, can provide isolation measures for close contact persons with different risk levels, can provide a convenient and efficient query tool for judging whether the close contact persons are the close contact persons or not by the public, fills the blank of risk probability estimation of the close contact persons in the public transportation field, and has important decision support significance for providing accurate research and judgment, joint prevention and control, public emotion guiding and emergency resource deployment for related departments.

Those of ordinary skill in the art will appreciate that: the drawing is a schematic diagram of one embodiment and the modules or flows in the drawing are not necessarily required to practice the invention.

From the above description of embodiments, it will be apparent to those skilled in the art that the present invention may be implemented in software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present invention.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus or system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, with reference to the description of method embodiments in part. The apparatus and system embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

The present invention is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present invention are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for identifying risk of cross infection of urban rail line-level intimate contact, comprising:

step S3, determining a RESFC set, a RETFC set and a REPFC set of cross-infection risk exposure trains according to microscopic travel behaviors and travel processes of the cases and travel physical paths, microscopic travel behaviors and travel processes of the closely connected persons, wherein the closely connected persons are closely connected persons at urban rail line level;

s4, constructing a cross infection risk exposure physical path RERFC set and an effective train ET set of the packer, giving a unique identification code to one piece of data, and constructing a packer initial data set Dataset (0) containing the case unique identification code;

Step S6, solving a passenger travel time probability distribution model based on an expectation maximization EM algorithm, and calculating the probability of a passing train of a passenger group and the passenger outbound travel time distribution parameters;

step S10, determining the cross infection risk level of the close-connected person according to the cross infection risk probability POC of the close-connected person and a set cross infection risk level classification strategy;

the risk exposure line REL, the risk exposure physical path RER and the travel duration of the acquired case in step S1 specifically include:

the risk exposure physical path RER of the case is:

wherein ,indicating that patient p is +.>Arrival station is +.>Is a travel path of the vehicle; l (L) _m Representing all site sets on line l; l (L) _o 、l _d Representing a site on line l;

step S013, inquiring AFC data of an automatic ticket selling and checking system of passengers in a database by taking intelligent traffic card numbers and travel OD pairs of cases as indexes, and determining case records and corresponding travel moments wherein ,/>Determining the travel duration of the case according to the travel time>

In the step S2, microscopic trip behaviors and trip time of the case are analyzed according to REL, RER and trip time of the case, and the method specifically includes:

In the step S3, the reset set, the refc set and the refc are determined according to the microscopic trip behavior and trip process of the case, and the trip physical path, the microscopic trip behavior and trip process of the packer, which specifically includes:

(2) The minimum outbound running time constraint of the terminal is that the arrival time of the train at the terminal is smaller than the outbound card swiping time:

step S033, determining REPFC, T according to the automatic train positioning AVL data of each train in the RETFC set ^repfc ＝[T ^start ,T ^end ]The following conditions are satisfied:

wherein ,representing train t ¹ Origin station l on REL in the same direction as case p ₁ Arrival time,/->Indicating the originating station l ₁ Maximum inbound travel time of (2); />Representing train t ^t Terminal l on REL in the same direction as case p _n At the moment of departure from the station,indicating terminal station l _n Maximum inbound travel time of (2);

step S4 is to construct a cross infection risk exposure physical path RERFC set and an effective train ET set of the packer, and specifically comprises the following steps:

step S041, determining RERFC set

Step S043, imitating the RETFC set constructing method involved in the step S032, and constructing the effective train ET set of the packer according to the code of the passenger OD pair and the trip time and combining the arrival time and the departure time of the trains at the O and D sites in the automatic train positioning data;

in the step S5, a passenger travel time probability distribution model of each station in the risk exposure line REL containing hidden variables is constructed based on the initial Dataset (0), and specifically includes:

(1) Conversion of the same OD to passenger population

The card entering and swiping time of the jth personal observation data in the kth passenger group isThe card swiping time of the outbound is->

(2) The packer initial Dataset (0) is partitioned into N classes,

N＝T ^end -T ^start /Δt (9)

(1)UD ^k Unobserved data representing class k passenger population

wherein ,

(4) The initial data set of the packer is satisfied by the complete data setThe method comprises the following steps:

(1) Constructing parameters to be estimated for a kth passenger population

wherein ,representing the probability of boarding the ith train for the kth passenger group, < >>Representing site l _s The inner passenger outbound time distribution parameters;

t ⁱⁿ Finger-sealer arrival time, t ^out Finger tight jointThe outbound time of the person, UD refers to unobserved information, and theta is a model to-be-solved parameter;

step S056, converting likelihood function into log likelihood function

Step S057, maximizing log likelihood function

L(θ)＝maxL _c (θ)＝max(logP(t ⁱⁿ ,t ^out ,UD|θ)) (20)；

The step S6 of solving the passenger travel time probability distribution model based on the expectation maximization EM algorithm, and calculating the passenger group journey-ascending train probability and the passenger outbound travel time distribution parameter specifically includes:

Step S061, determining the initial value θ=θ of the parameter ⁽⁰⁾ ；

Q(θ)＝E(L _c (θ)|(t ⁱⁿ ,t ^out ),θ ^(step) ) (21)

wherein ,θ^(step) Is the estimated value of step theta of the algorithm;

Wherein, the probability of the passing train of the kth passenger group in step +1 is thatCalculating passenger outbound time distribution parameters by using optimization method of L-BFGS-B>

in the step S7, the distribution parameters of the passenger arrival and departure time of each station, the cross infection risk probability of the trains in the RETFC set, and the packer train arrival probability are calculated according to the arrival and departure train probability of the passenger population and the distribution parameters of the passenger arrival and departure time, and specifically include:

Step S073, determining each station l of the risk exposure line according to the outbound running time distribution of the passengers and the inbound running time distribution curve of the station _s Passenger arrival time distribution parameters of (c)

Distinguishing cases from the packer according to the unique identification codes, and identifying the cross infection risk probability and the packer train routing probability of the trains in the RETFC set;

the updating the cross-infection risk exposure period reffc in step S8, updating the database (0) to the cross-infected person risk data set database (new), the unique identification code is unchanged, specifically includes:

exposing each station l of the line according to the risk _s The passenger arrival and departure time distribution parameter and the calculation formula T of REPFC ^repfc ＝[T ^start ,T ^end ]Updating REPFC, shortening the minimum inbound card swiping time and the maximum outbound card swiping time of Dataset (0) based on the updated time length of REPFC, and updating a cross-infected person risk data set Dataset (new), wherein the total time length of Dataset (new) is smaller than Dataset (0);

the calculating the risk probability of cross infection POC of the close contact person based on the risk Dataset of cross-infection person (new) in step S9 specifically includes:

wherein ,case p showing belonging to the kth passenger group can be booked for the train +.>The maximum range of k is N types, the train serial number is i, and the maximum range is t;

wherein ,indicating that the packer j belonging to the kth passenger group can log on the train +.>The maximum range of k is N types, the train serial number is i, and the maximum range is t;

2. A urban rail line level intimate contact cross-infection risk identification device, comprising:

updating a cross infection risk exposure period REPFC based on the cross infection risk probability of the trains in the RETFC set and the train trip probability of the close connector, updating a Dataset (0) into a cross infection risk data set Dataset (new) based on the updated REPFC, and keeping the unique identification code unchanged; calculating a cross-infection risk probability POC of the fitter based on a cross-infected person risk Dataset (new);

the risk level query module is used for determining the cross infection risk level of the close-connected person according to the POC and the set cross infection risk level classification strategy and providing a public query function;

The specific processing procedure for acquiring the risk exposure line REL, the risk exposure physical path RER and the travel duration of the case by the information acquisition module comprises the following steps:

the risk exposure physical path RER of the case is:

The information acquisition module analyzes the microscopic trip behaviors and trip time of the case according to REL, RER and trip time of the case, and the specific processing process comprises the following steps:

the data analysis and processing module is used for determining an ESFC set, a RETFC set and REPFC according to the microscopic trip behaviors and trip processes of the cases and trip physical paths, the microscopic trip behaviors and trip processes of the close-connected persons, wherein the specific processing process comprises the following steps:

the data analysis and processing module constructs a cross infection risk exposure physical path RERFC set and an effective train ET set of the packer, and the specific processing process comprises the following steps:

Step S041, determining RERFC set

the model construction and calculation module comprises a model construction module and a model calculation module;

(1) Conversion of the same OD to passenger population

(2) The packer initial Dataset (0) is partitioned into N classes,

N＝T ^end -T ^start /Δt (9)

(1)UD ^k unobserved data representing class k passenger population

wherein ,

(1) Constructing parameters to be estimated for a kth passenger population

t ⁱⁿ Finger-sealer arrival time, t ^out Indicating the outbound time of the packer, UD indicating unobserved information, and theta being the model to-be-solved parameter;

step S056, converting likelihood function into log likelihood function

Step S057, maximizing log likelihood function

L(θ)＝max L _c (θ)＝max(logP(t ⁱⁿ ,t ^out ,UD|θ)) (20)；

The model construction and calculation module is used for solving a passenger travel time probability distribution model based on an Expectation Maximization (EM) algorithm, calculating distribution parameters of the passenger group on-journey train probability and the passenger outbound travel time, and the specific processing process comprises the following steps:

step S061, determining the initial value θ=θ of the parameter ⁽⁰⁾ ；

Q(θ)＝E(L _c (θ)|(t ⁱⁿ ,t ^out ),θ ^(step) ) (21)

wherein ,θ^(step) Is the estimated value of step theta of the algorithm;

The model construction and calculation module calculates the distribution parameters of the passenger arrival and departure running time of each station according to the distribution parameters of the arrival and departure running time of the passenger group, and the cross infection risk probability of the trains in the RETFC set and the arrival and departure running probability of the close-connector trains, and the specific processing process comprises the following steps:

the model construction and calculation module updates REPFC in a cross infection risk exposure period, updates Dataset (0) into a cross infection person risk data set Dataset (new), and the unique identification code is unchanged, and the specific processing process comprises the following steps:

Exposing each station l of the line according to the risk _s The passenger arrival and departure time distribution parameter and the calculation formula T of REPFC ^repfc ＝[T ^start ,T ^end ]Updating REPFC based on updated RThe time length of the EPFC is shortened, the minimum inbound card swiping time and the maximum outbound card swiping time of the Dataset (0) are shortened, the cross-infection risk data set Dataset (new) is updated, and the total length of the Dataset (new) time is smaller than the Dataset (0);

the model construction and calculation module is also used for executing the following processing procedures:

wherein ,indicating that the packer j belonging to the kth passenger group can log on to train ti in RETFC set _p The maximum range of k is N types, the train serial number is i, and the maximum range is t;

/>