CN115964864A - Emergency control method for simulation-driven highway accident - Google Patents
Emergency control method for simulation-driven highway accident Download PDFInfo
- Publication number
- CN115964864A CN115964864A CN202211583521.5A CN202211583521A CN115964864A CN 115964864 A CN115964864 A CN 115964864A CN 202211583521 A CN202211583521 A CN 202211583521A CN 115964864 A CN115964864 A CN 115964864A
- Authority
- CN
- China
- Prior art keywords
- expressway
- control
- event
- highway
- accident
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention provides an emergency control method for an accident on a highway, which comprises the following steps: s1, establishing an expressway basic information base and an expressway accident information base, and constructing and verifying a basic simulation model; s2, establishing an expressway management and control strategy expert base and a simulation management and control scheme; s3, when an expressway accident occurs, analyzing the time and space influence range of the accident on line; s4, dynamically matching a traffic control strategy based on the highway control strategy expert base; and S5, after the emergency control of the expressway is finished, arranging the event information and the emergency management strategy information, and storing the event information and the emergency management strategy information as a new traffic control strategy sample into an expressway control strategy expert database. The invention has strong comprehensiveness and wide applicability, can reduce the influence of the expressway emergency on the traffic efficiency and improve the traffic safety situation.
Description
Technical Field
The invention relates to the technical field of traffic control of expressways, in particular to an emergency control method for simulation-driven expressway accidents.
Background
In recent years, the highway driving mileage in China is increased year by year and is stable in the world. However, the life and property safety of people is threatened by accidents such as multiple traffic accidents, severe weather, natural disasters, vehicle mechanical faults, dangerous chemical leakage and the like on the expressway. If a proper control strategy is not adopted to ensure the operation of the traffic system when an accident happens on the expressway, secondary traffic safety accidents can be possibly induced by the accident, and the traffic transportation efficiency is obviously reduced.
The highway promotes the development of national social economy, the traffic safety and the operation efficiency of the highway are concerned widely, and the emergency traffic control under the accident of the highway is not neglected. The highway traffic emergency management program stipulates that emergency response is divided into four levels according to the time and space range of traffic interruption influenced by accidents, and management measures are stipulated by classification. Traffic control strategies under the accident of the expressway comprise ramp control, variable speed limit, variable lane, toll station control, path induction, signal control of a road section connected with the expressway and the like, and the method is difficult to select the appropriate traffic control strategy quickly and accurately and is the emergency management of the expressway. A large number of scholars at home and abroad have studied the problem of dealing with unexpected events on the highway from the aspects of risk early warning, obstacle avoidance facilities, emergency rescue and the like, but the following two problems are not considered:
(1) Emergency control strategies under various different accidents on the highway are less considered, if the conventional research often limits the accidents to traffic control under modes of traffic jam, dangerous goods leakage, severe weather and the like, in order to expand the applicability of the invention, the invention is used for researching the various accidents and ensuring the accuracy and feasibility of emergency measures under different accidents;
(2) The mixed setting condition of various management and control strategies of the highway under a certain accident is less considered, for example, according to the structural characteristics of different highways and the grade of the accident, one or more management and control strategies are adopted to improve the traffic efficiency and the traffic safety situation.
Disclosure of Invention
The present invention aims to provide an emergency control method for simulation-driven highway accidents, so as to solve the defects of the prior art.
In order to achieve the purpose of the invention, the emergency control method for the simulation-driven expressway accident provided by the invention comprises the following steps of:
s1, establishing an expressway basic information base and an expressway accident information base, and constructing and verifying a basic simulation model;
s2, establishing an expressway management and control strategy expert base and a simulation management and control scheme;
s3, when an expressway accident occurs, analyzing the time and space influence range of the accident on line;
s4, dynamically matching a traffic control strategy based on an expressway control strategy expert base;
and S5, after the emergency control of the expressway is finished, arranging the event information and the emergency management strategy information, and storing the event information and the emergency management strategy information as a new traffic control strategy sample to an expressway control strategy expert library.
The step S1 includes:
s101, establishing a national highway basic information base and a highway accident information base;
s102, respectively simulating road conditions of a main line, an interweaving area, a ramp, a toll station and a road section connected with the ground of the expressway based on a variable cellular transmission model;
s103, selecting an actual expressway, splicing the simulated road condition graph and the actual expressway graph by adjusting the variable attributes of an expressway main line, an interweaving area, a ramp, a toll station, and a road junction or a road section which is connected with the ground, and operating each basic simulation model to obtain an average travel speed and a traffic volume evaluation index;
and S104, checking each basic simulation model according to the average travel speed and traffic data of the expressway, executing the step S2 if the checking requirement is met, otherwise, adjusting parameters of the simulation models, obtaining evaluation indexes again and executing the step S103.
The step of checking whether each basic simulation model meets the requirements comprises the following steps:
in each basic simulation model, the average hourly traffic volume of the expressway is input, and the average travel speed output in 1 hour is simulated
Number of "and" passing vehicles Q VCTM ' Highway average travel speed>
'and' highway average hour traffic Q h Respectively comparing, and if the following formula is met, considering that the simulation model meets the requirements;
further, the step S2 includes:
s201, establishing an expressway management and control strategy expert base;
s202, setting different traffic control strategies based on the basic simulation model of the S1, forming a new simulation model, numbering and storing the new simulation model for matching and calling in the step S4, and setting parameters of the simulation control scheme as hyper-parameters by the simulation model so as to rapidly simulate the traffic control strategies under different parameters and obtain evaluation indexes.
Further, the hyper-parameters set by different traffic control policies include:
the ramp control over parameters comprise the cell number of the ramp, the ramp flow input and output rate, the ramp control starting time and the ramp control duration;
the variable speed limit super parameter comprises a variable speed limit control starting cell, a variable speed limit control ending cell, a variable speed limit value, speed limit control starting time and speed limit control ending time;
the variable lane super parameter comprises a variable lane control starting cell, a variable lane control ending cell, a variable lane driving direction, variable lane control starting time and variable lane control ending time;
the toll station control super parameters comprise a cell number where the toll station is located, the open number of an ETC channel of the toll station, the open number of an artificial channel of the toll station, the open number of an ETC/artificial mixed channel of the toll station, the control start time of the toll station and the control end time of the toll station;
the path induction hyper-parameter comprises a cell number of an induction starting position, a cell number of an induction ending position, the traffic flow of the highway after the path induction is implemented, the traffic flow of a new path after the path induction is implemented, the induction control starting time and the induction control duration;
the signal control super-parameter of the lower ramp junction comprises a cell number controlled by a signal, a period controlled by the signal, the number of phases, a phase sequence and effective green light time of each phase;
the parameter solving mode of the traffic flow of the expressway after the implementation of the path induction and the traffic flow of the new path after the implementation of the path induction is as follows:
according to the BPR impedance function and the Wardrop first principle of traffic flow distribution, the path induction of the path induction is implemented to realize the traffic flow q of the expressway a And the traffic flow q of the new path after the path induction is implemented b The following equation is used in combination:
q a +q b ＝Q h +Q hnew (4)
in the formula: t is t a For inducing free flow travel time, t, from start to end of highway route b Free flow transit time for new path from induction start to end section, q a 、q b Traffic flow for the highway and new route after route guidance implementation, respectively, C a 、C b The practical traffic capacity of the expressway and the new path is respectively, alpha and beta are parameters, and the Federal road administration recommends alpha =0.15 and beta =4,Q hnew is The route induces traffic flow of the new route before implementation.
Further, the contents of the expert database of highway management and control policies include, but are not limited to: the method comprises the steps of emergency management strategy numbering, emergency management strategy type, emergency management strategy content description, management strategy starting time, management strategy ending time, longitude and latitude of a management strategy implementation place, strategy implementation effect, management strategy associated accident numbering and management strategy associated expressway numbering.
Further, the online analysis of the temporal and spatial impact range of the accident event comprises:
positioning the cells of the expressway according to the data of the expressway accident information base, and taking the cells where the accident is located as a center for analyzing the influence range;
drawing an average travel vehicle speed data curve for each cell, and counting data of three hours before and after an event occurs and three hours of historical data of 7 days, 14 days, 21 days in the same week and in the same time period;
setting a duration threshold for the outlier;
analyzing abnormal values by an IQR-based method;
selecting all the cells influenced by the event and the influence degree thereof;
establishing an event influence range prediction model based on Bayes, inputting characteristics of daily average traffic volume of a highway, design speed, average travel speed of the highway, number of lanes on the road, types of unexpected events, event start time, event end time, longitude and latitude of event places, number of cells where the events are located, length of the researched cells from the event, event grade, number of event blocking lanes and number of the highway where the events are associated, and labeling the influence degree of the cells by the events;
according to the attribute of the real-time event, the predicted event duration and the event start time are summed to replace the event end time, the influence degree of the cell on the event is combined, and the influence degree of all the cells on the accident event, namely the influence range of the event, can be obtained through multiple prediction and finishing.
The step S3 includes:
s301, when an accident happens to the expressway, acquiring historical detector data and current internet statistical data of the expressway from one month;
s302, analyzing the duration of the event based on a survival analysis model;
and S303, analyzing the influence range of the event based on an abnormal value identification method of the IQR method.
The survival analysis model analyzes the probability that the duration of an event represents that the duration of the event T is longer than the time T, and is obtained by the following formula:
wherein F (x) and F (T) respectively represent a density function and a distribution function of the event duration T, T is a continuous random variable, and S (T) is a survival function and is also an integral of the probability density function F (x).
The step S4 includes:
s401, training based on an expressway management and control strategy expert base to obtain an emergency strategy selection model;
s402, selecting a primary emergency management strategy type for the highway accident through an emergency strategy selection model;
s403, selecting secondary emergency management strategy real-time parameters for the highway accident through the basic simulation model;
and S404, outputting parameters corresponding to the optimal evaluation indexes in the basic simulation model as a detailed management and control scheme of the expressway accident.
Compared with the prior art, the invention has the beneficial effects that at least:
(1) The invention has strong comprehensiveness and wide applicability. The emergency strategy control method is a mixed strategy control method, integrates key technologies such as event impact analysis, control strategy simulation and scheme quantitative automatic selection, can form a complete highway emergency strategy control system, can adapt to highway networks under different scenes, and has high practicability.
(2) The types of accidents and traffic emergency management and control strategies considered by the method are various and comprehensive, the types of the considered expressways are abundant along with the increase of the sample library, and compared with the prior method for adopting a certain management and control strategy for expressway traffic emergency according to experience or singly, the method is more scientific and reliable.
(3) According to the emergency strategy control method, the optimal management and control method of the highway under different accident scenes is determined through quantitative evaluation indexes, so that social and economic losses caused by accidents can be reduced to the maximum extent, and the safety of traffic users and the operation of a traffic system are guaranteed.
Drawings
FIG. 1 is a flow chart of a method for emergency control of simulation driven highway accidents in accordance with the present invention;
FIG. 2 is an example of a simulation model of a highway based on VCTM in the emergency policy control method provided by the present invention;
FIG. 3 is a flow chart of the analysis of the impact range of the highway accident in the emergency strategy control method provided by the present invention;
FIG. 4 is an exemplary graph of an average travel speed data curve of a certain cell on a highway according to the emergency strategy control method provided by the present invention;
fig. 5 is an example of a hybrid policy control method under a highway traffic accident scenario in the emergency policy control method provided by the present invention.
Detailed Description
In order to more clearly illustrate the present invention, the present invention is further described in detail below with reference to the following examples and the accompanying drawings. It is to be understood by persons skilled in the art that the following detailed description is illustrative and not restrictive, and is not to be taken as limiting the scope of the invention.
The invention provides a simulation-driven emergency control method for highway accidents, which is based on the condition that traffic is not completely interrupted but traffic capacity is influenced under the highway accidents, adopts a proper traffic control strategy and determines a specific implementation scheme of the strategy by evaluating the time and space influence range of the accidents, and provides a simulation-driven emergency control method for highway accidents.
The invention researches the condition that an expressway is influenced by an accident and needs traffic control but the traffic is not completely interrupted, firstly, an expressway basic information base, an expressway accident information base and an accident emergency management strategy information base (namely an expressway control strategy expert base) are established, and information data are integrated as samples; when an accident occurs, analyzing the time and space influence range of the expressway management and control strategy expert base simulation model prediction event by combining real-time internet data; determining a traffic control strategy and a specific implementation scheme thereof by an emergency strategy selection model and a simulation model based on Variable Cellular Transmission Model (VCTM); and finally, recording the accident and the control strategy thereof to an information base to form a new sample.
Accidents considered by the invention include traffic accidents, severe weather, natural disasters, mechanical faults of vehicles and dangerous chemical leakage; the considered traffic emergency control strategy comprises ramp control, variable speed limit, variable lane, toll station control, path induction and signal control; the types of the considered expressways are complex and various and are perfected with the supplement of the information base. The emergency strategy control method forms a complete expressway accident response system, has wide applicability, and can reduce the influence of accidents on traffic transportation efficiency and reduce social and economic benefits loss caused by accidents.
Referring to fig. 1 and 5, the emergency control method for simulation driven highway accident provided by the present invention includes the following steps:
s1, establishing an expressway basic information base and an expressway accident information base, and constructing and verifying a basic simulation model.
In this embodiment, based on the basic information base of the highway and the unexpected event information base of the highway, the simulation model components of the main line, the intersection area, the ramp, the toll station and the ground connection road section are first established based on the cellular transmission model, and then the simulation model components modify parameters and are repeatedly used to assemble a simulation model capable of simulating the traffic flow operation state of the complete highway, which is called as a basic simulation model, wherein the step S1 includes but is not limited to:
s101, establishing a national highway basic information base and a highway accident information base.
In one embodiment of the present invention, the contents of the national highway base information base include, but are not limited to: the highway geometric linear form comprises highway serial numbers, names, highway starting point longitude and latitude, highway finishing point longitude and latitude, highway daily average traffic, highway average hour traffic, highway hour traffic (24 groups in total from 0 hour to 23 hours), design speed, highway average travel speed, lane number and highway geometric linear form.
The geometrical line shape of the highway is stored in the form of an AutoCAD file, and the longitude and latitude of the starting point of the highway in the file are converted into plane coordinates and are superposed with the origin coordinates of the AutoCAD.
In one embodiment of the present invention, the contents of the highway contingency information base include, but are not limited to: the method comprises the following steps of accident numbering, accident type, accident starting time, accident ending time, accident site longitude and latitude, length of an accident influence road section, accident grade, number of accident blocking lanes and accident-related highway numbering.
After the 'longitude and latitude of the event place' is converted into the plane coordinate, the 'number of the expressway associated with the event' can be associated with the expressway and the specific position thereof when the accident happens.
S102, respectively simulating five road conditions of a main line of the expressway, an interweaving area, a ramp, a toll station and a road section connected with the ground based on the variable cellular transmission model.
In one embodiment of the present invention, the Variable Cell Transmission Model (VCTM) is a model that divides a highroad into several segments of different lengths, called cells, and disperses the time into uniform time segments, and the traffic flow status of the cells is updated once per time segment.
Optionally, in the variable cellular transmission model, attributes of a main line and a ramp of the high-speed highway, such as the number of lanes, lane width, road section length, road starting point coordinates, road ending point coordinates, road section speed limit, road surface type (such as cement and asphalt), whether an emergency stop lane exists or not, altitude, gradient and turning radius can be modified; the properties of the expressway intersection area, such as the number of lanes, the length of a road section, the coordinates of a road starting point, the coordinates of a road ending point, the speed limit of the road section, the type of a road surface (such as cement and asphalt), whether an emergency stop lane exists or not, the altitude, the gradient, the turning radius, the width of each lane, the lane-to-lane direction change direction of each lane and the type of a road (main line-main line, main line-ramp, ramp-main line and ramp-ramp) connected with each lane can be modified; the highway toll station has the characteristics that the number of toll channels, the length of a road section, the coordinates of a road starting point, the coordinates of a road ending point, the types of the toll channels (such as ETC toll collection and manual toll collection) and the working efficiency attribute of each toll channel can be modified; the highway and ground connection section has the characteristics that the number of lanes, the lane width, the length of the section, the coordinates of the starting point of the road, the coordinates of the ending point of the road, the speed limit of the section, the gradient, the channelizing scheme of a first intersection and the communication control scheme of the first intersection can be modified.
The recording mode of the first intersection canalization scheme and the signal control scheme of the expressway ground connection road section is as follows: taking an entrance lane where a road section connected with the expressway is positioned as an initial serial number, and taking the clockwise direction as serial numbers of other entrance lanes at the intersection; further numbering each lane of the entrance road by starting with the lane close to the center line of the road, and recording the direction of the entrance road to which each lane leads; and recording the start and end time of the green light, the start and end time of the yellow light and the start and end time of the red light of each lane under the same signal timing scheme.
S103, selecting an actual expressway, splicing the simulated road condition graph and the actual expressway graph by adjusting the variable attributes of an expressway main line, an interweaving area, a ramp, a toll station and a road junction or a road section which is connected with the ground, and operating each basic simulation model to obtain average travel speed and traffic volume evaluation indexes.
In the embodiment of the invention, a real complete expressway is selected, a simulation model (namely a national expressway basic information base) of the expressway is fitted as much as possible after splicing by adjusting the variable attributes of an expressway main line, an interweaving area, a ramp, a toll station and a road junction or a road section which is connected with the ground, and the simulation model is operated to obtain the average travel speed and traffic volume evaluation index.
The complete highway can be regarded as comprising a main line, an interweaving area, a ramp, a toll station, and a road junction or a road section which is connected with the ground, and step S101 is to establish simulation model components of the main line, the interweaving area, the ramp, the toll station and the road section which is connected with the ground based on a cellular transmission model. And then modifying parameters by the simulation model component for repeated use through the step S102 to combine into a complete highway simulation model. For example, if a practical and complete expressway consists of a main line 1, an intersection zone 1, a ramp 2, a ground connection section 1 and a ground connection section 2, a simulation model of the main line, the intersection zone, the ramp and the ground connection section is needed. And establishing a basic simulation model connection relation according to the vehicle flow direction, for example, if the vehicle needs to flow from the main line 1 to the interweaving area 1, the main line 1 and the interweaving area 1 are spliced. The description above shows which parts are selected and combined to form a complete simulation model, after verification, different management and control strategies are set for the simulation model through step S203, and step S4 is used for calling. The invention can improve the efficiency and avoid modeling each highway again by multiplexing the basic simulation model.
And S104, checking the simulation model according to the average travel speed and traffic data of the expressway, entering S2 if the checking requirement is met, and otherwise entering S103 to adjust model parameters and obtain the evaluation index again.
In one embodiment of the invention, the imitation isInputting 'average hourly traffic volume of expressway' into the true model, and simulating 'average travel speed' output in 1 hour
Number of "and" passing vehicles Q VCTM 'Highway average travel speed based on' and highway basic information base>
'and' average hourly traffic volume Q of highway h "compare respectively, if satisfy equation (1), equation (2) simultaneously, consider that basic simulation model satisfies the requirement.
S2, establishing an expressway management and control strategy expert base and a simulation management and control scheme.
In this embodiment, the step S2 includes:
s201, establishing an expressway management and control strategy expert base.
In one embodiment of the present invention, the contents of the expert database of highway management policies include, but are not limited to: the method comprises the steps of emergency management strategy numbering, emergency management strategy type, emergency management strategy content description, management strategy starting time, management strategy ending time, longitude and latitude of a management strategy implementation place, strategy implementation effect, management strategy associated accident numbering and management strategy associated expressway numbering.
After the longitude and latitude of the implementation place of the management strategy are converted into plane coordinates, the management strategy can be associated to the highway where the management strategy is located and the specific position of the highway through the high-speed number associated with the management strategy.
In one embodiment of the invention, each highway may correspond to a plurality of different emergency management strategies.
Each highway may correspond to a plurality of emergency management strategies, the emergency management strategies comprise emergency management strategy types and management strategy real-time selectable parameters, wherein the emergency management strategy types can be regarded as primary strategies and are divided into ramp control, variable speed limit, variable lanes, toll station control, path induction and lower ramp junction signal control, the management strategy real-time selectable parameters can be regarded as secondary strategies, the parameters in the ramp control are flow input and output rates of ramp control, the parameters in the variable speed limit are variable speed limit values, the parameters in the variable lanes are variable lane driving directions, the parameters in the toll station control are opening numbers of various toll channel types, the parameters in the path induction are induction paths and vehicle induction ratios, and the parameters in the lower ramp junction signal control are designed as junction phase and timing schemes.
S202, setting different control strategies based on the simulation model of the S1, forming a new simulation model, numbering and storing the new simulation model so as to facilitate calling of the S4, and setting parameters of a control scheme to be hyper-parameters by the simulation model so as to rapidly simulate the control scheme under different parameters and obtain evaluation indexes.
In one embodiment of the present invention, a highway simulation model established based on VCTM is shown in fig. 2, in which a complete highway is simulated in the embodiment, and the highway simulation model is composed of 26 cells, wherein cells 1 and 16 are starting cells and represent starting points at two ends of the highway; the cells 15 and 26 are termination cells and represent end points at both ends of the highway; cells 5 and 9 are ramp cells and represent places where ramp control can be implemented; the cells 11 to 14 are rate-limiting cells, which means that the sections between the cells 11 to 14 can implement variable rate-limiting control; the cells 17 and 25 are toll station cells and represent that toll station control strategies can be implemented to determine the number of open or closed lanes; the cells 17 to 25 are lane-changeable cells, and indicate that the section between the cells 17 to 25 can implement lane-changeable control; the cells 10, 28 and 29 are signal control cells; the other cells are common cells.
In one embodiment of the present invention, the superparameters set by different management and control schemes are different: 1) The ramp control super-parameter comprises a cellular number where the ramp is located, a ramp flow input and output rate, ramp control starting time and ramp control duration. 2) The variable speed limit super-parameter comprises a variable speed limit control starting cellular, a variable speed limit control ending cellular, a variable speed limit value, speed limit control starting time and speed limit control ending time. 3) The variable lane crossing parameter includes a variable lane control start cell, a variable lane control end cell, a variable lane driving direction, a variable lane control start time, and a variable lane control end time. 4) The toll station control super-parameter comprises a cell number of a toll station, the opening number of an ETC channel of the toll station, the opening number of an artificial channel of the toll station, the opening number of an ETC/artificial mixed channel of the toll station, the control starting time of the toll station and the control ending time of the toll station. 5) The path induction hyper-parameter comprises a cell number of an induction starting position, a cell number of an induction ending position, the traffic flow of the highway after the path induction is implemented, the traffic flow of a new path after the path induction is implemented, the induction control starting time and the induction control duration. 6) The signal control over-parameters of the junction of the lower ramp include the cell number controlled by the signal, the period, the phase number, the phase sequence and the effective green time of each phase controlled by the signal.
In one embodiment of the present invention, each management and control scheme may be set at multiple locations of the highway, and different management and control schemes may be set in combination; when the expressway does not have the implementation condition of a certain management and control scheme, the management and control scheme is in an unselected state; the parameter solving mode of the traffic flow of the expressway after the route guidance implementation and the traffic flow of the new route after the route guidance implementation is as follows:
according to the BPR impedance function and the Wardrop (the Wardrop balance principle, the scholars Wardrop puts forward the first principle and the second principle of the traffic network balance definition and lays the foundation of traffic flow distribution) first principle, the path induction a Traffic flow q of new route after 'and' route induction implementation b "obtained by the formula (3) and the formula (4) in combination:
q a +q b ＝Q h +Q hnew (4)
in the formula: t is t a For inducing free flow travel time, t, from beginning to end of highway path b Free flow transit time for new path from induction start to end section, q a 、q b Traffic flow for the highway and new route after route guidance implementation, respectively, C a 、C b The practical traffic capacity of the expressway and the new path is respectively, alpha and beta are parameters, and the Federal road administration recommends alpha =0.15 and beta =4,Q hnew Traffic flow of the new path before implementation is induced for the path.
And S3, when the expressway accidents occur, analyzing the time and space influence ranges of the accidents on line.
In this embodiment, the step S3 includes:
s301, when an accident happens to the expressway, historical detector data and current internet statistical data of the expressway from one month to the current are obtained.
In one embodiment of the invention, when an accident event occurs on the expressway, the method is triggered to acquire the historical number of vehicles and the average travel vehicle speed data from one month to the current on the expressway, wherein the vehicle number data can be acquired by an induction coil vehicle detector or a video vehicle detector arranged on the expressway; the average travel speed data is crawled by map software at regular time, in the embodiment, each small section of the crawled data is set according to the cells of the highway simulation model at intervals of 5 minutes, and each cell is associated with each small section to obtain the average travel speed data of each cell of the highway as the basis of S302.
S302, analyzing the duration of the event based on the survival analysis model.
The survival Analysis model is an Analysis method introduced from the field of medical statistics (Press c. Statistical and ecometric Methods for Transportation Data Analysis, second Edition [ M ]. CRC Press, 2010.), and has strong capability in processing deleted Data, influencing variable Analysis and the like due to the research on the duration of events in a targeted manner.
In one embodiment of the present invention, the survival analysis of the duration of the highway accident event refers to a method for analyzing and deducing the duration of the event according to the existing data, and researching the relationship between the duration of the event and a plurality of influencing factors and the magnitude of the influence thereof. Deleted data refers to data that is truncated for various reasons and is one of the most important characteristics of survival analysis; the survival function (survival rate) represents the probability that the event duration T is longer than the time T, denoted as S (T):
wherein F (x) and F (T) respectively represent a density function and a distribution function of the event duration T. T is a continuous random variable, and the survival function is also the integral of the probability density function f (x).
The risk function (conditional death probability) represents the probability of the end of a subsequent time interval Δ t after the duration t of the accident, denoted h (t):
h (t) represents the probability of the end of a subsequent time interval Δ t after the contingency duration t; t is the duration of the study; f (t) is the probability that the duration value is t; s (t) is survival rate, and can be obtained by an expression, the greater the value of the risk function which is taken as conditional probability, the greater the probability that the event is ended at the next moment is.
In one embodiment of the present invention, lifetime map analysis, kaplan-Mayer analysis (Kaplan Meier is a one-way survival analysis), cox regression analysis are used as a set of survival analysis methods to study the duration of the unexpected event. Wherein the lifetime map is used for recording and counting data, recording population to be analyzedState change of the event, statistics representing statistics such as median and the like; kaplan-Mayer analysis is a nonparametric method for estimating survival and risk functions, noting T 1 ＜T 2 …T n For samples representing n event durations, the estimation of the survival function S (t) by Kaplan-Mayer analysis is:
in the formula, T to tal is the total number of events,
is the ith event after sorting. Where event i needs to be complete non-punctured data, the punctured data is recorded separately. Kaplan-Mayer can quantitatively analyze the influence degree of a certain influence factor on the event duration, and judge whether the influence of the factor on the result is obviously different by combining Log-rank test.
The Cox regression analysis is a semi-parametric regression model, and the ratio of the risk function to the basic risk function is taken as a dependent variable to reflect the influence of different independent variables. Expressed as:
h(t,x)＝h 0 (t)exp(β 1 x 1 +β 2 x 2 +…+β i x i ) (8)
in the formula, beta 1 ,β 2 …β i Is a regression coefficient, x 1 ,x 2 …x i Is an independent variable, h (t, x) is a hazard function, h 0 (t) is a base hazard function, representing the hazard function inherent to the duration of the event in the absence of other factors.
Where parametric regression is a form of a model assumed in advance, and then the data is used to estimate the coefficients of this model. Nonparametric regression is a model fitting directly from data, without assuming a model form. Semi-parametric regression is that some of the structures in the model are known, parameters need to be estimated, and some of the structures are unknown. cox regression belongs to a semi-parametric regression model.
In one embodiment of the invention, the regression coefficient of the hazard function of the duration of the highway accident is obtained through data analysis of a 'sample information base', and the duration of the accident can be predicted according to the input argument value of the real-time accident. The sample information base comprises road section basic information, accident information and management and control strategy information. The 'sample information base' refers to a national highway basic information base, a highway accident information base and a highway management and control strategy expert base.
And S303, analyzing the influence range of the event based on an abnormal value identification method of the IQR method.
In one embodiment of the invention, a research idea of a traffic wave model is combined, based on "historical one month of each cell of the expressway to current average travel vehicle speed data" acquired in S3, a determined event influence range analysis process is shown in FIG. 3, and the overall idea is to determine an influence range of a sample event, establish a prediction model of the event influence range, and input characteristics of a real-time event to predict the influence range of the real-time event.
Step 1, positioning the cells of the expressway according to the data of the expressway accident information base, and taking the cells where the accident is located as the center of the analysis of the influence range; step 2, drawing an average travel vehicle speed data curve for each cell, specifically counting data of three hours before and after an event occurs and data of three hours after history of 7 days, 14 days, 21 days, the same week and the same time period, where the average travel vehicle speed data curve of a certain cell drawn in this embodiment is shown in fig. 4, and obtaining speed-record point broken line graphs of each road section history and accident time period after once exponential smoothing according to historical data of three hours before and after an accident occurs (13: 00-16:00 three hours, namely 180 minutes, of each data observation time, and the ordinate is the average travel speed in the cells; step 3, setting a duration threshold of an abnormal value according to professional experience, wherein the duration threshold is set to be 10 minutes, namely if the average travel time of a certain cell exceeds the duration of the abnormal value threshold by 10 minutes compared with the history, the cell is considered to be influenced by an event; step 4, abnormal value analysis, namely an abnormal value analysis method based on an IQR (InterQuartileRange) method is adopted in the invention; step 5, selecting all cells influenced by the event and the influence degree thereof, wherein the influence degree is represented by the average travel speed reduction degree; step 6, establishing an event influence range prediction model based on Bayes, inputting the characteristics of daily average traffic volume of an expressway, design speed, average highway travel speed, number of lanes on a road, type of an accident event, event starting time, event ending time, longitude and latitude of an event place, the number of cells where the event is located, the length of the cells where the researched cells are away from the event, event grade, the number of event blocking lanes and the number of expressway related to the event, and labeling the influence degree of the cells on the event (0 is not influenced, a positive value is influenced positively, the average travel speed is increased, a negative value is influenced negatively and the average travel speed is decreased); and 7, according to the attribute of the real-time event, summing the predicted event duration and the event start time to replace the event end time, outputting a result of the influence degree of the cells on the event by combining other characteristics, and obtaining the influence degree of all the cells on the unexpected event, namely the influence range of the event, through multiple prediction and finishing.
And S4, dynamically matching the traffic control strategy based on the highway control strategy expert base.
In this embodiment, the step S4 specifically includes:
s401, training based on the highway management and control strategy expert base to obtain an emergency strategy selection model.
S402, selecting a primary emergency management strategy type for the highway accident through the emergency strategy selection model.
In one embodiment of the invention, the emergency strategy selection model is obtained by training the sample information base data of S2, and the model can be based on Bayes, random forest and other machine learning methods and can also be based on a neural network. The method is characterized by comprising the steps of inputting the number of an expressway, daily average traffic volume of the expressway, average hour traffic volume of the expressway, design speed, average travel speed of the expressway, number of road lanes, types of unexpected events, event starting time, event ending time, longitude and latitude of event places, length of event influence road sections, event grade, number of event blocking lanes and number of expressway associated with the events, outputting an 'emergency management strategy' (1 is ramp control, 2 is variable speed limit, 3 is variable lane, 4 is toll station control, 5 is path induction and 6 is lower ramp junction signal control), and inputting information from the known expressway, real-time unexpected event attributes and predicted event duration time and influence range when a control scheme is selected for a real-time event.
The emergency strategy selection model can be based on machine learning methods such as Bayes and random forests, and can also be based on a neural network. The method is characterized by comprising the steps of inputting the number of an expressway, daily average traffic volume of the expressway, average hour traffic volume of the expressway, design speed, average travel speed of the expressway, number of road lanes, types of unexpected events, event starting time, event ending time, longitude and latitude of event places, length of event influence road sections, event grade, number of event blocking lanes and number of expressway associated with the events, outputting an 'emergency management strategy' (1 is ramp control, 2 is variable speed limit, 3 is variable lane, 4 is toll station control, 5 is path induction and 6 is lower ramp junction signal control), and inputting information from the known expressway, real-time unexpected event attributes and predicted event duration time and influence range when a control scheme is selected for a real-time event.
And S403, selecting secondary emergency management strategy real-time parameters for the highway accident through the simulation model.
In one embodiment of the invention, after the primary emergency management policy type is selected for the event, the simulation model corresponding to the step S2 is called, and the control variable changes the hyper-parameter of the simulation model management and control scheme. And outputting evaluation indexes of 'average travel speed', 'passing vehicle number' and 'vehicle average delay' from the beginning of the simulation to the end of the event for each set of parameters, wherein because some parameters are continuous variables, the evaluation indexes are completed in an interpolation mode after upper and lower limit simulation groups are set for the variables.
And S404, outputting parameters corresponding to the optimal evaluation indexes in the basic simulation model as a detailed management and control scheme of the expressway accident.
In one embodiment of the invention, the total delay time of the vehicle from the start of the event simulation to the end of the event is calculated for each set of parameters, the total delay time of the vehicle is the product of the number of passing vehicles and the average delay of the vehicles, and when the product is minimum, the evaluation index of the set of parameters is considered to be optimal.
And S5, after the emergency control of the expressway is finished, arranging the event information and the emergency management strategy information, and storing the event information and the emergency management strategy information as a new traffic control strategy sample into an expressway control strategy expert database.
In one embodiment of the invention, the system reports that a link rear-end collision happens to a certain highway, so that abnormal congestion is caused. An example of an event handling process is as follows: step 1, the event triggering system acquires the historical number of vehicles and the average travel speed data of the expressway from one month to the current; step 2, analyzing the duration of the event based on a survival analysis model, and analyzing the influence range of the event based on an abnormal value identification method of an IQR method; step 3, combining an expressway basic information base and an expressway accident information base, taking expressway information, traffic accident attributes and predicted event duration and influence range as input, and selecting a first-level emergency management strategy type for the event through an emergency strategy selection model, wherein the first-level emergency management strategy type is '1-ramp control, 5-path induction and 6-down-ramp junction signal control'; step 4, calling a simulation model which is trained off line and comprises a management and control strategy of 1 ramp control, 5 path induction and 6 ramp-down junction signal control, wherein in a secondary strategy, the cell number of an over-parameter ramp controlled by a ramp is set as [ the cell number of the ramp on an event point ], the ramp flow input rate is set as [ the lane is not closed and 1 lane is closed ], the ramp control starting time is set as [ the fastest time for emergency managers to reach the ramp on the ramp ], the ramp control duration is set as [10 minutes, 20 minutes, 30 minutes, 40 minutes ], the cell number of the over-parameter induction starting position of the path induction is set as [ the cell number of the nearest intersection of the ramp on the event point ], the cell number of an induction ending position is set as [ the cell number of the nearest intersection of the next ramp on the event point ], the traffic flow of the expressway after the path induction implementation [ 50% of the original expressway flow), 40%,30%,20%,10% ], induction control starting time [ fastest time for implementing induction control ], induction control duration [10 minutes, 20 minutes, 30 minutes, 40 minutes ], the cell number of the lower ramp junction signal control over-parameter under signal control is set as [ the cell number included in the lower ramp junction ], the period, the phase number and the phase sequence of signal control are consistent with the original scheme, the effective green light time of the road section of the lower ramp is set as [30 seconds, 40 seconds, 50 seconds ], and the over-parameter of the simulation model control scheme is variably changed by controlling. Outputting evaluation indexes of 'average travel speed', 'passing vehicle number' and 'vehicle average delay' from the beginning of the event simulation to the end of the event for each set of parameters; step 5, calculating the total delay time of the vehicles from the beginning of the simulation to the end of the event for each set of parameters, wherein the total delay time of the vehicles is the product of the number of passing vehicles and the average delay of the vehicles, and when the product is minimum, the evaluation index of the set of parameters is considered to be optimal; and 6, sorting the event information and the emergency management strategy information, and counting the event information and the emergency management strategy information as a new sample to an information base.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. An emergency control method for simulation driven expressway accidents is characterized by comprising the following steps:
s1, establishing an expressway basic information base and an expressway accident information base, and constructing and verifying a basic simulation model;
s2, establishing an expressway management and control strategy expert base and a simulation management and control scheme;
s3, when an expressway accident occurs, analyzing the time and space influence range of the accident on line;
s4, dynamically matching a traffic control strategy based on an expressway control strategy expert base;
and S5, after the emergency control of the expressway is finished, arranging the event information and the emergency management strategy information, and storing the event information and the emergency management strategy information as a new traffic control strategy sample into an expressway control strategy expert database.
2. The emergency control method for simulating a driven expressway accident according to claim 1, wherein the step S1 comprises:
s101, establishing a national highway basic information base and a highway accident information base;
s102, respectively simulating road conditions of a main line, an interweaving area, a ramp, a toll station and a road section connected with the ground of the expressway based on a variable cellular transmission model;
s103, selecting an actual expressway, splicing the simulated road condition graph and the actual expressway graph by adjusting the variable attributes of an expressway main line, an interweaving area, a ramp, a toll station, and a road junction or a road section which is connected with the ground, and operating each basic simulation model to obtain an average travel speed and a traffic volume evaluation index;
and S104, checking each basic simulation model according to the average travel speed and traffic data of the expressway, executing the step S2 if the checking requirement is met, otherwise, adjusting parameters of the simulation models, obtaining evaluation indexes again and executing the step S103.
3. The method of emergency control of simulation driven highway accidents according to claim 2, wherein verifying whether each basic simulation model satisfies the requirements comprises:
in each basic simulation model, the average hourly traffic volume of the expressway is input, and the average travel speed output in 1 hour is simulated
Number of "and" passing vehicles Q VCTM 'Highway average travel speed based on' and highway basic information base>
'and' average hourly traffic volume Q of highway h Respectively comparing, and if the following formula is met, the simulation model is considered to meet the requirements;
4. the emergency control method for simulating an accident of a driven highway according to claim 1, wherein the step S2 comprises:
s201, establishing an expressway management and control strategy expert base;
s202, setting different traffic control strategies based on the basic simulation model of the S1, forming a new simulation model and numbering and storing the new simulation model for matching and calling in the step S4, and meanwhile setting parameters of the simulation control scheme as hyper-parameters by the simulation model for rapidly simulating the traffic control strategies under different parameters and obtaining evaluation indexes.
5. The method of claim 4, wherein the hyper-parameters set by different traffic control policies comprise:
the ramp control over parameters comprise the cell number of the ramp, the ramp flow input and output rate, the ramp control starting time and the ramp control duration;
the variable speed limit super-parameter comprises a variable speed limit control starting cellular, a variable speed limit control ending cellular, a variable speed limit value, speed limit control starting time and speed limit control ending time;
the variable lane super parameter comprises a variable lane control starting cell, a variable lane control ending cell, a variable lane driving direction, variable lane control starting time and variable lane control ending time;
the toll station control super-parameter comprises a cell number where the toll station is located, the open number of an ETC channel of the toll station, the open number of an artificial channel of the toll station, the open number of an ETC/artificial mixed channel of the toll station, the control start time of the toll station and the control end time of the toll station;
the path induction hyper-parameter comprises a cell number of an induction starting position, a cell number of an induction ending position, the traffic flow of the highway after the path induction is implemented, the traffic flow of a new path after the path induction is implemented, the induction control starting time and the induction control duration;
the signal control super-parameter of the lower ramp junction comprises a cell number controlled by a signal, a period controlled by the signal, the number of phases, a phase sequence and effective green light time of each phase;
the parameter solving mode of the traffic flow of the expressway after the route guidance implementation and the traffic flow of the new route after the route guidance implementation is as follows:
according to the BPR impedance function and the Wardrop first principle of traffic flow distribution, the path induction of the path induction is implemented to realize the traffic flow q of the expressway a And the traffic flow q of the new path after the path induction is implemented b The following equation is used to obtain:
q a +q b ＝Q h +Q hnew (4)
in the formula：t a For inducing free flow travel time, t, from start to end of highway route b Free flow transit time for new path from induction start to end section, q a 、q b Traffic flow for the highway and new route after route guidance implementation, respectively, C a 、C b The practical traffic capacity of the expressway and the new path is respectively, alpha and beta are parameters, and the Federal road administration recommends alpha =0.15 and beta =4,Q hnew Traffic flow of the new path before implementation is induced for the path.
6. The method of emergency control of simulation driven highway accidents according to claim 2, wherein the contents of said highway management strategy expert base include but are not limited to: the method comprises the steps of emergency management strategy numbering, emergency management strategy type, emergency management strategy content description, management strategy starting time, management strategy ending time, longitude and latitude of a management strategy implementation place, strategy implementation effect, management strategy associated accident numbering and management strategy associated expressway numbering.
7. The method of claim 1, wherein the online analysis of temporal and spatial impact ranges of accidents comprises:
positioning the cells of the expressway according to the data of the expressway accident information base, and taking the cells where the accident is positioned as a center for analyzing the influence range;
drawing an average travel vehicle speed data curve for each cell, and counting data of three hours before and after an event occurs and three hours of historical data of 7 days, 14 days, 21 days in the same week and in the same time period;
setting a duration threshold for the outlier;
analyzing abnormal values by an IQR-based method;
selecting all the cells influenced by the event and the influence degree thereof;
establishing an event influence range prediction model based on Bayes, inputting characteristics of daily average traffic volume of a highway, design speed, average travel speed of the highway, number of lanes on the road, types of unexpected events, event start time, event end time, longitude and latitude of event places, number of cells where the events are located, length of the researched cells from the event, event grade, number of event blocking lanes and number of the highway where the events are associated, and labeling the influence degree of the cells by the events;
according to the attribute of the real-time event, the predicted event duration and the event start time are summed to replace the event end time, the influence degree of the cell on the event is combined, and the influence degree of all the cells on the accident event, namely the influence range of the event, can be obtained through multiple prediction and finishing.
8. The emergency control method for simulating a driven expressway accident according to claim 1, wherein the step S3 comprises:
s301, when an accident happens to the expressway, acquiring historical detector data and current internet statistical data of the expressway from one month;
s302, analyzing the duration of the event based on a survival analysis model;
and S303, analyzing the influence range of the event based on an abnormal value identification method of the IQR method.
9. The method of claim 8, wherein the survival analysis model analyzes the probability that the duration of the event is longer than T, and the probability is obtained by the following formula:
wherein F (x) and F (T) respectively represent a density function and a distribution function of the event duration T, T is a continuous random variable, and S (T) is a survival function and is also an integral of the probability density function F (x).
10. The emergency control method for simulating a driven expressway accident according to claim 1, wherein the step S4 comprises:
s401, training based on an expressway management and control strategy expert base to obtain an emergency strategy selection model;
s402, selecting a first-level emergency management strategy type for the expressway accident through an emergency strategy selection model;
s403, selecting real-time parameters of a secondary emergency management strategy for the expressway accident through the basic simulation model;
and S404, outputting parameters corresponding to the optimal evaluation indexes in the basic simulation model as a detailed management and control scheme of the expressway accident.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211583521.5A CN115964864A (en) | 2022-12-09 | 2022-12-09 | Emergency control method for simulation-driven highway accident |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211583521.5A CN115964864A (en) | 2022-12-09 | 2022-12-09 | Emergency control method for simulation-driven highway accident |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115964864A true CN115964864A (en) | 2023-04-14 |
Family
ID=87362627
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211583521.5A Pending CN115964864A (en) | 2022-12-09 | 2022-12-09 | Emergency control method for simulation-driven highway accident |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115964864A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117690301A (en) * | 2024-02-04 | 2024-03-12 | 福建省高速公路科技创新研究院有限公司 | Expressway diversion induction strategy considering induction compliance rate |
-
2022
- 2022-12-09 CN CN202211583521.5A patent/CN115964864A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117690301A (en) * | 2024-02-04 | 2024-03-12 | 福建省高速公路科技创新研究院有限公司 | Expressway diversion induction strategy considering induction compliance rate |
| CN117690301B (en) * | 2024-02-04 | 2024-04-23 | 福建省高速公路科技创新研究院有限公司 | Expressway diversion induction method considering induction compliance rate |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deveci et al. | Fuzzy Power Heronian function based CoCoSo method for the advantage prioritization of autonomous vehicles in real-time traffic management | |
| CN106205156B (en) | A kind of intersection self-healing combination control method for the mutation of part lane flow | |
| CN104157139B (en) | A kind of traffic congestion Forecasting Methodology and method for visualizing | |
| CN103473469B (en) | A kind of sector traffic situation multi_levels fuzzy evaluation method based on objective indicator | |
| CN104778834B (en) | Urban road traffic jam judging method based on vehicle GPS data | |
| CN109272745A (en) | A kind of track of vehicle prediction technique based on deep neural network | |
| CN109118787A (en) | A kind of car speed prediction technique based on deep neural network | |
| CN113674522A (en) | Ramp control method for solving problem of congestion of bottleneck road section of expressway | |
| CN115964864A (en) | Emergency control method for simulation-driven highway accident | |
| Wu et al. | Simulation and evaluation of speed and lane-changing advisory of CAVS at work zones in heterogeneous traffic flow | |
| CN116597642A (en) | Traffic jam condition prediction method and system | |
| Yang et al. | A hybrid methodology for freeway work-zone optimization with time constraints | |
| Akbar et al. | Methodology for simulating heterogeneous traffic flow at intercity roads in developing countries: a case study of university road in peshawar | |
| Zhong et al. | Deep Q‐Learning Network Model for Optimizing Transit Bus Priority at Multiphase Traffic Signal Controlled Intersection | |
| Sharma et al. | Evaluating ASCT Operations for Dodge Street Corridor | |
| McLin et al. | Development and evaluation of a control system for regional traffic management | |
| Crowther | A Comparison of CORSIM and INTEGRATION for the Modeling of Stationary Bottlenecks | |
| Sohag et al. | Demand forecasting for a domestic airport-A case study | |
| Khatoon et al. | Binary probabilistic models for pedestrians’ crossing behaviour and risk at the Free Left Turn: Delhi, India | |
| Roustom | Environmental Impacts of Connected and Automated Vehicles Considering Traffic Flow and Road Characteristics | |
| Sharma | Integrated behavioral and economic framework for improving dilemma zone protection systems | |
| Yu et al. | Simulation-based safety analysis of diversion and confluence area in urban underground loop | |
| Ling et al. | Modelling of traffic capacity under traffic accident | |
| Abd-Allah et al. | Evaluation of Common Traffic Operation Softwares on the Basis of Relevancy to Roads Intersections. | |
| Kim et al. | Enhancement of freeway incident traffic management and resulting benefits. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |