CN108182802B - Traffic safety analysis method based on information attenuation model and driven by license plate data mining - Google Patents
Traffic safety analysis method based on information attenuation model and driven by license plate data mining Download PDFInfo
- Publication number
- CN108182802B CN108182802B CN201711429278.0A CN201711429278A CN108182802B CN 108182802 B CN108182802 B CN 108182802B CN 201711429278 A CN201711429278 A CN 201711429278A CN 108182802 B CN108182802 B CN 108182802B
- Authority
- CN
- China
- Prior art keywords
- traffic
- influence coefficient
- flow
- equal
- road section
- 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.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 19
- 238000007418 data mining Methods 0.000 title claims abstract description 15
- 238000011156 evaluation Methods 0.000 claims abstract description 56
- 206010039203 Road traffic accident Diseases 0.000 claims description 27
- 230000000630 rising effect Effects 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000007423 decrease Effects 0.000 claims description 9
- 238000011144 upstream manufacturing Methods 0.000 claims description 8
- 230000002238 attenuated effect Effects 0.000 claims description 4
- 230000006399 behavior Effects 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 15
- 230000000694 effects Effects 0.000 description 3
- 230000008014 freezing Effects 0.000 description 3
- 238000007710 freezing Methods 0.000 description 3
- 238000007433 macroscopic evaluation Methods 0.000 description 2
- 238000007431 microscopic evaluation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000012847 principal component analysis method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a traffic safety analysis method based on an information attenuation model and driven by license plate data mining, and particularly relates to the field of road traffic, wherein the traffic safety analysis method comprises the following steps: s01, selecting characteristic evaluation factors for evaluating traffic safety; and S02, quantifying each characteristic evaluation factor, and calculating the influence coefficient of each characteristic evaluation factor. The influence of historical illegal and accident factors on traffic safety is calculated by adopting an information attenuation model. And S03, comprehensively evaluating the traffic safety level of the urban road through the characteristic evaluation factor influence coefficients of a plurality of characteristics. The method has the advantages of real-time and dynamic evaluation of road traffic safety and simple comprehensive evaluation method.
Description
Technical Field
The invention relates to the field of road traffic, in particular to a traffic safety analysis method based on an information attenuation model and driven by license plate data mining.
Background
The urban road traffic safety evaluation aims at guaranteeing the safety of users of urban roads, starts with preventing traffic accidents and reducing accidents, carries out all-around evaluation on the urban roads, reveals potential risk factors of the urban roads when accidents happen, takes corresponding measures and guarantees the safety of urban road traffic. However, it is very complicated to evaluate the road traffic safety level in different areas, and different evaluation results may be obtained based on different indexes. The evaluation of the road traffic safety system is generally divided into macroscopic evaluation and microscopic evaluation, wherein the macroscopic evaluation comprises an absolute number method, an accident rate method, a model method, an accident intensity method, four-index relative numbers and the like, and the microscopic evaluation comprises a traffic accident rate method, an absolute number-accident rate method, a risk degree judgment, a Lobarov model and the like. Common comprehensive evaluation methods include a gray clustering evaluation method, a principal component analysis method, an analytic hierarchy process and the like.
However, the above methods are often difficult to dynamically evaluate urban road traffic safety in real time, and the comprehensive evaluation method is complex and difficult to apply in engineering. Therefore, it is an urgent need to solve the problem of developing a real-time, simple and universal traffic safety analysis method.
Disclosure of Invention
The invention provides a traffic safety analysis method based on an information attenuation model and driven by license plate data mining, and the traffic safety analysis method has the advantages of real-time and dynamic evaluation of road traffic safety and simple comprehensive evaluation method.
In order to achieve the purpose, the invention provides the following technical scheme:
a traffic safety analysis method based on an information attenuation model and driven by license plate data mining comprises the following steps:
s01, selecting characteristic evaluation factors for evaluating traffic safety;
and S02, quantifying each characteristic evaluation factor, and calculating the influence coefficient of each characteristic evaluation factor. The influence of historical illegal and accident factors on traffic safety is calculated by adopting an information attenuation model.
And S03, comprehensively evaluating the traffic safety level of the urban road through the influence coefficients of the characteristic evaluation factors of the plurality of characteristics.
The method is used for calculating the influence coefficient of the pre-selected characteristic evaluation factor in real time based on the license plate identification data, accident data and the like acquired in real time, and calculating the traffic safety level of the urban road by integrating various influence coefficients acquired through the information attenuation model. The evaluation method is based on the license plate identification data acquired in real time, so that the information source is accurate, the method is convenient to apply to the actual work of evaluating traffic safety, and the method can be applied in engineering.
Preferably, the characteristic evaluation factors selectable in the present invention include weather conditions, traffic flow conditions, traffic events, traffic violations, and historical traffic accidents.
Preferably, the traffic flow state comprises sudden change of flow, speed difference of road sections, proportion of road sections and vehicle types and headway; the traffic events include traffic accident events and road construction events.
According to timeliness of the data flow, the importance of information contained in the data flow is attenuated continuously along with time, so that characteristic evaluation factors distinguish a current event and a historical event, the current event needs to be calculated dynamically in real time, and the historical event needs to be calculated through a historical model. The influence coefficients of characteristic evaluation factors such as sudden change of flow, speed difference of road sections, proportion of vehicle types of road sections, headway and the like contained in the traffic flow state need to be dynamically calculated in real time.
Preferably, the step of dynamically calculating the influence coefficient of the flow rate mutation in real time comprises the following steps:
a1, calculating the rising amplitude of the sudden flow change;
real-time calculation is carried out according to a unit cycle, road section flow is calculated according to license plate identification data, and the difference between the road section flow of the current cycle and the road section flow of the previous unit cycle is larger than a threshold phi1Or the ratio of the traffic flow of the current period to the traffic flow of the previous period is larger than a threshold value phi2If so, generating sudden flow change or the flow deviates from the historical normal rule, and simultaneously calculating the rising amplitude of the sudden flow change;
b1, obtaining influence coefficients of different flow sudden changes corresponding to the rising amplitude according to calculation;
the rising amplitude is more than 0% and less than or equal to 15%, and the corresponding influence coefficient is 0.001;
the rising amplitude is more than 15 percent and less than or equal to 30 percent, and the corresponding influence coefficient is 0.125;
the rising amplitude is more than 30 percent and less than or equal to 45 percent, and the corresponding influence coefficient is 0.375;
the rising amplitude is more than 45% and less than or equal to 60%, and the corresponding influence coefficient is 0.625;
the rising amplitude is more than 60 percent, and the influence coefficient of the corresponding flow sudden change is 0.875. More specifically, the step of dynamically calculating the influence coefficient of the speed difference of the road section in real time comprises the following steps:
a2, calculating the speed difference value of the road section;
the real-time calculation is carried out in a unit cycle, the license plate number, the passing time and the position information of the vehicle are respectively acquired according to the upstream and downstream electric alarm bayonets of the road section with the known length, the same vehicle is determined through the matching of the upstream and downstream license plates, and the time required by the vehicle running is obtained through acquiring the passing time of the vehicle on the upstream and downstream electric alarm bayonets. And determining the distance traveled by the vehicle according to the position information of the vehicle, and obtaining the average speed of the road section as distance/time.
Calculating to obtain the average speed of the road section;
the speed difference value of the road section is the difference between the current period of the same road section and the average speed of the road section in the previous period;
b2, obtaining different influence coefficients corresponding to the speed difference of the road section according to calculation;
the speed difference value of the road section is more than 0km/h and less than or equal to 15km/h, and the corresponding influence coefficient is 0.001;
the speed difference value of the road section is more than 15km/h and less than or equal to 20km/h, and the corresponding influence coefficient is 0.125;
the speed difference value of the road section is more than 20km/h and less than or equal to 25km/h, and the corresponding influence coefficient is 0.375;
the speed difference value of the road section is more than 25km/h and less than or equal to 30km/h, and the corresponding influence coefficient is 0.625;
the speed difference value of the road section is more than 30km/h, and the corresponding influence coefficient is 0.875.
Preferably, the step of dynamically calculating the influence coefficient of the road section vehicle type proportion in real time comprises the following steps:
a3, calculating the ratio of the large vehicle in the daytime to the total vehicle flow in the daytime;
calculating in real time in a unit period, and obtaining the number of large vehicles in the daytime and the total flow in the daytime according to the vehicle type information acquired by the data of the electric police card port, so as to calculate the ratio of the large vehicles in the daytime to the total flow of the vehicles in the daytime;
b3, obtaining different influence coefficients corresponding to the ratio of the white-day large-sized vehicle to the daytime vehicle total flow according to calculation;
the ratio of the white-day large-sized vehicle to the total vehicle flow in the daytime is more than 0% and less than or equal to 20%, and the corresponding influence coefficient is 0.001;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 20% and less than or equal to 35%, and the corresponding influence coefficient is 0.125;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 35% and less than or equal to 50%, and the corresponding influence coefficient is 0.375;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 50% and less than or equal to 65%, and the corresponding influence coefficient is 0.625;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the day is more than 65%, and the corresponding influence coefficient is 0.875.
In determining the impact coefficients of traffic events:
the residual traffic capacity value of the road section with the traffic incident is more than 80 percent of the pre-designed traffic capacity value, and the influence coefficient of the corresponding traffic incident is 0.125;
the residual traffic capacity value of the road section with the traffic incident is more than 50 percent of the pre-designed traffic capacity value and less than or equal to 80 percent of the pre-designed traffic capacity value, and the influence coefficient of the corresponding traffic incident is 0.375;
the residual traffic capacity value of the road section with the traffic incident is less than or equal to 50 percent of the pre-designed traffic capacity value, and the influence coefficient of the corresponding traffic incident is 0.625;
when the remaining traffic capacity value of the traffic incident road section is 0, the influence coefficient of the corresponding traffic incident is 0.875.
In determining the impact coefficients of traffic law violation:
using the mode support degree included in the factor in the information attenuation model to change along with the time t and carrying out attenuation by using an attenuation function POWER (lambda, ti/TLi), wherein, lambda is larger than 0, then:
ai＝bi*POWER(0.5，ti/TLi)
wherein, aiIs shown at tiInfluence coefficient of total number of internal violation on traffic safety; TLiHalf-decay time; t is tiA time interval representing the time of occurrence of the most recent event from the current time; biRepresenting an initial coefficient;
at tiWhen the average number of illegal acts per month is less than or equal to 4 and greater than 0, the corresponding biIs 0.125;
at tiWhen the average number of illegal acts per month is less than or equal to 8 and greater than 4, the corresponding bi0.375;
at tiIn the interior of said container body,when the average number of illegal acts per month is less than or equal to 12 and greater than 8, the corresponding biIs 0.625;
at tiWhen the average number of illegal acts per month is greater than 12, the corresponding biIs 0.875.
In determining the impact factors of historical traffic accidents:
using the mode support degree included in the factor in the information attenuation model to change along with the time t and carrying out attenuation by using an attenuation function POWER (lambda, ti/TLi), wherein, lambda is larger than 0, then:
Ai＝Bi*POWER(λ，Ti/TLi)
wherein A isiIs shown at TiInfluence coefficient of traffic accident to traffic safety; TLiFor half-decay time, TiA time interval representing the time of occurrence of the most recent event from the current time; b isiThe initial coefficient is represented as a function of time,
at TiWhen the average monthly traffic accident frequency is more than 1 and less than or equal to 2, the corresponding BiIs 0.25;
at TiWhen the average monthly traffic accident frequency is more than 2 and less than or equal to 4, the corresponding BiIs 0.50;
at TiWhen the average monthly traffic accident frequency is more than 4 and less than or equal to 6, the corresponding BiIs 0.75;
at TiWhen the average monthly traffic accident frequency is more than 6, the corresponding BiIs 1.
Among the factors that influence the meteorological conditions:
meteorological conditions are generally divided into: fog, rain, wind, freezing, snow. If severe weather such as heavy rain, strong wind, heavy fog, freezing and the like occurs, a1 is 0.125, otherwise a1 is 0.001;
as a preferred batch, step S03 is specifically:
according to the influence coefficient of the one-way characteristic factor, if a certain item has no score, the influence coefficient is 0.001. When a plurality of feature evaluation factors are combined, the influence of a single feature evaluation factor increases, and as a result, (0, 1):
f＝min(max(ai)+avg(ai)，0.99)
wherein f represents an influence coefficient of a characteristic evaluation factor of a plurality of characteristics, aiA coefficient value representing each characteristic evaluation factor;
the link safety factor decreases as the influence coefficient of the characteristic evaluation factor increases, but the speed of the decrease decreases, and as a result, (0, 10):
S＝10*(1-EXP(-LOG(f，0.05)))
wherein S represents a traffic safety index for the current time slot link.
Dividing the traffic safety index into 4 traffic safety grades according to the traffic safety index, and establishing a corresponding evaluation set which is divided into { unsafe, general safe and safer } - { 0-2.5, 2.5-5, 5-7.5, 7.5-10 }.
The invention has the beneficial effects that:
the method is used for calculating the influence coefficient of the pre-selected characteristic evaluation factor in real time based on the license plate identification data, accident data and the like acquired in real time, and calculating the traffic safety level of the urban road by integrating various influence coefficients acquired through the information attenuation model. The evaluation method is based on the license plate identification data acquired in real time, so that the information source is accurate, the method is convenient to apply to the actual work of evaluating traffic safety, and the method can be applied in engineering.
Drawings
FIG. 1 is a table of the score values of the urban road traffic safety indexes in the embodiment 2;
Detailed Description
The embodiment provides a technical scheme:
example 1
A traffic safety analysis method based on an information attenuation model and driven by license plate data mining comprises the following steps:
s01, selecting weather conditions, traffic flow states, traffic incidents, traffic illegal behaviors and historical traffic accidents for evaluating characteristic evaluation factors of traffic safety; the traffic flow state comprises the following characteristic evaluation factors: flow rate sudden change, speed difference of road sections, vehicle type proportion of the road sections and vehicle head time distance; the traffic events include the following characteristic evaluation factors: traffic accident incidents and road construction incidents.
S02, calculating the influence coefficients of all the characteristic evaluation factors selected in the step S01 on the road traffic safety. See below, respectively:
calculating the influence coefficient of the sudden change of the flow on the road traffic safety: firstly, real-time calculation is carried out in a unit period, road section flow is calculated according to license plate identification data, and the difference between the road section flow in the current period and the road section flow in the previous unit period is larger than a threshold value phi1Or the ratio of the traffic flow of the current period to the traffic flow of the previous period is larger than a threshold value phi2If the flow rate suddenly changes or the flow rate deviates from the historical normal rule, the rising amplitude of the sudden flow rate change needs to be calculated. When the rising amplitude is more than 0% and less than or equal to 15%, the corresponding influence coefficient is 0.001, the rising amplitude is more than 15% and less than or equal to 30%, and the corresponding influence coefficient is 0.125; the rising amplitude is more than 30 percent and less than or equal to 45 percent, and the corresponding influence coefficient is 0.375; the rising amplitude is more than 45% and less than or equal to 60%, and the corresponding influence coefficient is 0.625; the rise amplitude is greater than 60%, and the corresponding influence coefficient is 0.875.
Calculating the influence coefficient of the speed difference of the road section on the traffic safety: the real-time calculation is carried out in a unit cycle, the license plate number, the passing time and the position information of the vehicle are respectively acquired according to the upstream and downstream electric alarm bayonets of the road section with the known length, the same vehicle is determined through the matching of the upstream and downstream license plates, and the time required by the vehicle running is obtained through acquiring the passing time of the vehicle on the upstream and downstream electric alarm bayonets. And determining the distance traveled by the vehicle according to the position information of the vehicle, and obtaining the average speed of the road section as distance/time. Thereby calculating the average speed of the road section. When the speed difference value of the road section is more than 0km/h and less than or equal to 15km/h, the corresponding influence coefficient is 0.001; the speed difference value of the road section is more than 15km/h and less than or equal to 20km/h, and the corresponding influence coefficient is 0.125; the speed difference value of the road section is more than 20km/h and less than or equal to 25km/h, and the corresponding influence coefficient is 0.375; the speed difference value of the road section is more than 25km/h and less than or equal to 30km/h, and the corresponding influence coefficient is 0.625; the speed difference value of the road section is more than 30km/h, and the corresponding influence coefficient is 0.875.
Calculating the influence coefficient of the vehicle type proportion on the road traffic safety: the real-time calculation is carried out in a unit cycle, and according to the vehicle type information acquired by the data of the electric police card port, the influence coefficient of the large-sized vehicle which runs in the daytime is calculated as an example: the number of the white-sky large-sized vehicles and the total daytime flow are calculated, and therefore the ratio of the white-sky large-sized vehicles to the total daytime vehicle flow is calculated. When the ratio of the large vehicle to the total vehicle flow in the daytime is more than 0% and less than or equal to 20%, the corresponding influence coefficient is 0.001; the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 20% and less than or equal to 35%, and the corresponding influence coefficient is 0.125; the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 35% and less than or equal to 50%, and the corresponding influence coefficient is 0.375; the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 50% and less than or equal to 65%, and the corresponding influence coefficient is 0.625; the ratio of the white-day large-sized vehicles to the total vehicle flow in the day is more than 65%, and the corresponding influence coefficient is 0.875.
Calculating the influence coefficient of the traffic event on the road safety: the influence of traffic incident factors on road traffic safety is analyzed, and the traffic incident generally comprises traffic accidents, vehicle faults, scattered objects and other incidents influencing road traffic capacity. When the residual traffic capacity value of the road section with the traffic incident is more than 50 percent of the pre-designed traffic capacity value and less than or equal to 80 percent of the pre-designed traffic capacity value, the influence coefficient of the corresponding traffic incident is 0.375; the residual traffic capacity value of the road section with the traffic incident is less than or equal to 50 percent of the pre-designed traffic capacity value, and the influence coefficient of the corresponding traffic incident is 0.625; when the remaining traffic capacity value of the traffic incident road section is 0, the influence coefficient of the corresponding traffic incident is 0.875.
Calculating the influence coefficient of the traffic violation behaviors on road safety: because the traffic violation belongs to the historical event, and the historical event needs to be calculated through the historical model, the mode support degree included by the traffic violation factor in the information attenuation model changes with the time t and is attenuated by the attenuation function POWER (λ, ti/TLi), wherein λ > 0, then:
ai＝bi*POWER(0.5，ti/TLi)
wherein, aiIs shown at tiInfluence coefficient of total number of internal violation on traffic safety; TLiHalf-decay time; t is tiA time interval representing the time of occurrence of the most recent event from the current time; biRepresenting the initial coefficients. The initial coefficient values are divided into the following cases: at tiWhen the average number of illegal acts per month is less than or equal to 4 and greater than 0, the corresponding biIs 0.125; at tiWhen the average number of illegal acts per month is less than or equal to 8 and greater than 4, the corresponding bi0.375; at tiWhen the average number of illegal acts per month is less than or equal to 12 and greater than 8, the corresponding biIs 0.625; at tiWhen the average number of illegal acts per month is greater than 12, the corresponding biIs 0.875. The total number of illegal activities can be counted for nearly 3 months, and the average number of illegal activities per month can be obtained.
Calculating the influence coefficient of the traffic accident on the road safety: since the traffic accident belongs to a historical event, and the historical event needs to be calculated through a historical model, the mode support degree included by the factor in the information attenuation model changes along with the time t and is attenuated by an attenuation function POWER (lambda, ti/TLi), wherein lambda is greater than 0, then:
Ai＝Bi*POWER(λ，Ti/TLi)
wherein A isiIs shown at TiInfluence coefficient of traffic accident to traffic safety; TLiFor half-decay time, TiA time interval representing the time of occurrence of the most recent event from the current time; b isiRepresenting the initial coefficients. The initial coefficient values are divided into the following cases: at TiWhen the average monthly traffic accident frequency is more than 1 and less than or equal to 2, the corresponding BiIs 0.25; at TiWhen the average monthly traffic accident frequency is more than 2 and less than or equal to 4, the corresponding BiIs 0.5; at TiWhen the average monthly traffic accident frequency is more than 4 and less than or equal to 6, the corresponding BiIs 0.75; at TiAverage number of monthly traffic accidentsWhen the number is greater than 6, corresponding BiIs 1. The total number of illegal activities can be counted for data of nearly 3 months, and the average number of traffic accidents per month can be obtained.
Calculating the influence coefficient of the meteorological condition road safety: if severe weather such as heavy rain, strong wind, heavy fog, freezing and the like occurs, a1 is 0.125, otherwise a1 is 0.001;
and S03, according to the influence coefficient of the one-way characteristic factor, if any item has no score, the influence coefficient is 0.001. When a plurality of feature evaluation factors are combined, the influence of a single feature evaluation factor increases, and as a result, (0, 1):
f＝min(max(ai)+avg(ai)，0.99)；
wherein f represents an influence coefficient of a characteristic evaluation factor of a plurality of characteristics, aiRepresenting the influence coefficient corresponding to each characteristic evaluation factor;
the link safety factor decreases as the influence coefficient of the characteristic evaluation factor increases, but the speed of the decrease decreases, and as a result, (0, 10):
S＝10*(1-EXP(-LOG(f，0.05)))
wherein S represents a traffic safety index of a road segment at the current time period, aiA coefficient value representing each characteristic evaluation factor;
dividing the traffic safety index into 4 traffic safety grades according to the traffic safety index, and establishing a corresponding evaluation set which is divided into { unsafe, general safe and safer } - { 0-2.5, 2.5-5, 5-7.5, 7.5-10 }.
Example 2
In this example, the steps described in example 1 were applied to perform a specific analysis on traffic safety.
As shown in fig. 1, in combination with the influence weight and the score of the road traffic safety evaluation index, when a traffic incident occurs in the current road segment in the period, and a certain degree of sudden change in traffic and a certain degree of vehicle speed difference occur in the road segment, and the number of illegal events of the road segment in one month of history is 7, the number of accidents is 4, the road traffic safety score can be found to be 0.276, and the safety level in the current period is unsafe.
Claims (8)
1. A traffic safety analysis method based on an information attenuation model and driven by license plate data mining is characterized by comprising the following steps:
s01, selecting characteristic evaluation factors for evaluating traffic safety;
s02, quantifying each characteristic evaluation factor, and calculating the influence coefficient of each characteristic evaluation factor;
s03, comprehensively evaluating the traffic safety level of the urban road through the influence coefficients of the characteristic evaluation factors;
wherein the characteristic evaluation factors comprise meteorological conditions, traffic flow states, traffic events, traffic illegal behaviors and historical traffic accidents;
meanwhile, the mode support degree included by the traffic violation behavior factor in the information attenuation model is changed along with the time t and is attenuated by an attenuation function POWER (lambda, ti/TLi), wherein lambda is larger than 0, then:
ai＝bi*POWER(0.5，ti/TLi)
wherein, aiIs shown at tiInfluence coefficient of total number of internal violation on traffic safety; TLiHalf-decay time; t is tiA time interval representing the time of occurrence of the most recent event from the current time; biRepresenting an initial coefficient;
at tiWhen the average number of illegal acts per month is less than or equal to 4 and greater than 0, the corresponding biIs 0.125;
at tiWhen the average number of illegal acts per month is less than or equal to 8 and greater than 4, the corresponding bi0.375;
at tiWhen the average number of illegal acts per month is less than or equal to 12 and greater than 8, the corresponding biIs 0.625;
at tiWhen the average number of illegal acts per month is greater than 12, the corresponding biIs 0.875.
2. The traffic safety analysis method based on the information attenuation model and driven by license plate data mining according to claim 1, wherein the traffic flow state comprises sudden flow change, speed difference of road sections, proportion of vehicle types of road sections and headway; the traffic events include traffic accident events and road construction events.
3. The traffic safety analysis method based on the information attenuation model and driven by license plate data mining according to claim 2, characterized in that the real-time dynamic calculation of the influence coefficient of the sudden change of the flow comprises the following steps:
a1, calculating the rising amplitude of the sudden flow change;
real-time calculation is carried out according to a unit cycle, road section flow is calculated according to license plate identification data, and the difference between the road section flow of the current cycle and the road section flow of the previous unit cycle is larger than a threshold phi1Or the ratio of the traffic flow of the current period to the traffic flow of the previous period is larger than a threshold value phi2If so, generating flow sudden change, and simultaneously calculating the rising amplitude of the flow sudden change;
b1, obtaining influence coefficients of different flow sudden changes corresponding to the rising amplitude according to calculation;
the rising amplitude is more than 0% and less than or equal to 15%, and the corresponding influence coefficient is 0.001;
the rising amplitude is more than 15 percent and less than or equal to 30 percent, and the corresponding influence coefficient is 0.125;
the rising amplitude is more than 30 percent and less than or equal to 45 percent, and the corresponding influence coefficient is 0.375;
the rising amplitude is more than 45% and less than or equal to 60%, and the corresponding influence coefficient is 0.625;
the rising amplitude is more than 60 percent, and the influence coefficient of the corresponding flow sudden change is 0.875.
4. The traffic safety analysis method based on the information attenuation model and driven by license plate data mining as claimed in claim 2, wherein the step of dynamically calculating the influence coefficient of the speed difference of the road section in real time comprises the following steps:
a2, calculating the speed difference value of the road section;
calculating in real time in unit cycle, respectively acquiring license plate numbers, passing time and vehicle position information according to upstream and downstream electric alarm checkpoints of a road section with known length, and calculating to obtain the average speed of the road section through upstream and downstream license plate matching;
the speed difference value of the road section is the difference between the current period of the same road section and the average speed of the road section in the previous period;
b2, obtaining different influence coefficients corresponding to the speed difference of the road section according to calculation;
the speed difference value of the road section is more than 0km/h and less than or equal to 15km/h, and the corresponding influence coefficient is 0.001;
the speed difference value of the road section is more than 15km/h and less than or equal to 20km/h, and the corresponding influence coefficient is 0.125;
the speed difference value of the road section is more than 20km/h and less than or equal to 25km/h, and the corresponding influence coefficient is 0.375;
the speed difference value of the road section is more than 25km/h and less than or equal to 30km/h, and the corresponding influence coefficient is 0.625;
the speed difference value of the road section is more than 30km/h, and the corresponding influence coefficient is 0.875.
5. The traffic safety analysis method based on the information attenuation model and driven by license plate data mining as claimed in claim 2, wherein the step of dynamically calculating the influence coefficient of the road section vehicle type proportion in real time comprises the following steps:
a3, calculating the ratio of the large vehicle in the daytime to the total vehicle flow in the daytime;
calculating in real time in a unit period, and obtaining the number of large vehicles in the daytime and the total flow in the daytime according to the vehicle type information acquired by the data of the electric police card port, so as to calculate the ratio of the large vehicles in the daytime to the total flow of the vehicles in the daytime;
b3, obtaining different influence coefficients corresponding to the ratio of the white-day large-sized vehicle to the daytime vehicle total flow according to calculation;
the ratio of the white-day large-sized vehicle to the total vehicle flow in the daytime is more than 0% and less than or equal to 20%, and the corresponding influence coefficient is 0.001;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 20% and less than or equal to 35%, and the corresponding influence coefficient is 0.125;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 35% and less than or equal to 50%, and the corresponding influence coefficient is 0.375;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the daytime is more than 50% and less than or equal to 65%, and the corresponding influence coefficient is 0.625;
the ratio of the white-day large-sized vehicles to the total vehicle flow in the day is more than 65%, and the corresponding influence coefficient is 0.875.
6. The traffic safety analysis method based on the information attenuation model and driven by license plate data mining according to claim 1, characterized in that:
the residual traffic capacity value of the road section with the traffic incident is more than 80 percent of the pre-designed traffic capacity value, and the influence coefficient of the corresponding traffic incident is 0.125;
the residual traffic capacity value of the road section with the traffic incident is more than 50 percent of the pre-designed traffic capacity value and less than or equal to 80 percent of the pre-designed traffic capacity value, and the influence coefficient of the corresponding traffic incident is 0.375;
the residual traffic capacity value of the road section with the traffic incident is less than or equal to 50 percent of the pre-designed traffic capacity value, and the influence coefficient of the corresponding traffic incident is 0.625;
when the remaining traffic capacity value of the traffic incident road section is 0, the influence coefficient of the corresponding traffic incident is 0.875.
7. The traffic safety analysis method based on the information attenuation model and driven by license plate data mining according to claim 1, characterized in that:
using the mode support degree included in the factor in the information attenuation model to change along with the time t and carrying out attenuation by using an attenuation function POWER (lambda, ti/TLi), wherein, lambda is larger than 0, then:
Ai＝Bi*POWER(λ，Ti/TLi)
wherein A isiIs shown at TiInfluence coefficient of traffic accident to traffic safety; TLiFor half-decay time, TiA time interval representing the time of occurrence of the most recent event from the current time; b isiThe initial coefficient is represented as a function of time,
at TiWhen the average monthly total number of traffic accidents is more than 1 and less than or equal to 2, the corresponding BiIs 0.25;
at TiWhen the average monthly total number of traffic accidents is more than 2 and less than or equal to 4, the corresponding BiIs 0.5;
at TiWhen the average monthly total number of traffic accidents is more than 4 and less than or equal to 6, the corresponding BiIs 0.75;
at TiWhen the average monthly total number of traffic accidents is more than 6, the corresponding BiIs 1.
8. The traffic safety analysis method based on the information attenuation model and driven by license plate data mining according to claim 1, characterized in that:
step S03 specifically includes:
according to the influence coefficient of the one-way characteristic factor, if a certain item has no score, the influence coefficient is 0.001; when a plurality of feature evaluation factors are combined, the influence of a single feature evaluation factor increases, and as a result, (0, 1):
f＝min(max(ai)+avg(ai)，0.99)
wherein f represents an influence coefficient of a characteristic evaluation factor of a plurality of characteristics, aiRepresenting the influence coefficient corresponding to each characteristic evaluation factor;
the link safety factor decreases as the influence coefficient of the characteristic evaluation factor increases, but the speed of the decrease decreases, and as a result, (0, 10):
S＝10*(1-EXP(-LOG(f，0.05)))
wherein S represents a traffic safety index of a road segment at the current time period, aiA coefficient value representing each characteristic evaluation factor;
dividing the traffic safety index into 4 traffic safety grades according to the traffic safety index, and establishing a corresponding evaluation set which is divided into { unsafe, general safe and safer } - { 0-2.5, 2.5-5, 5-7.5, 7.5-10 }.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711429278.0A CN108182802B (en) | 2017-12-26 | 2017-12-26 | Traffic safety analysis method based on information attenuation model and driven by license plate data mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711429278.0A CN108182802B (en) | 2017-12-26 | 2017-12-26 | Traffic safety analysis method based on information attenuation model and driven by license plate data mining |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108182802A CN108182802A (en) | 2018-06-19 |
| CN108182802B true CN108182802B (en) | 2020-10-16 |
Family
ID=62547582
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711429278.0A Active CN108182802B (en) | 2017-12-26 | 2017-12-26 | Traffic safety analysis method based on information attenuation model and driven by license plate data mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108182802B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118770A (en) * | 2018-09-11 | 2019-01-01 | 东南大学 | A kind of road section capacity method for digging based on Traffic monitoring data |
| CN112037513B (en) * | 2020-09-01 | 2023-04-18 | 清华大学 | Real-time traffic safety index dynamic comprehensive evaluation system and construction method thereof |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101465059A (en) * | 2008-12-31 | 2009-06-24 | 公安部交通管理科学研究所 | Early-warning system for inspecting and judging urban road traffic safety situation |
| CN102231231A (en) * | 2011-06-16 | 2011-11-02 | 同济大学 | Area road network traffic safety situation early warning system and method thereof |
| CN102368354A (en) * | 2011-10-19 | 2012-03-07 | 北京航空航天大学 | Road security evaluation method based on floating vehicle data acquisition |
| US20140358326A1 (en) * | 2005-06-01 | 2014-12-04 | Allstate Insurance Company | Motor vehicle operating data collection and analysis |
| CN104978853A (en) * | 2014-04-01 | 2015-10-14 | 中国移动通信集团公司 | Road traffic safety evaluating method and system |
| CN105608902A (en) * | 2016-03-28 | 2016-05-25 | 辽宁省交通科学研究院 | Expressway black spot identification system and method |
-
2017
- 2017-12-26 CN CN201711429278.0A patent/CN108182802B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140358326A1 (en) * | 2005-06-01 | 2014-12-04 | Allstate Insurance Company | Motor vehicle operating data collection and analysis |
| CN101465059A (en) * | 2008-12-31 | 2009-06-24 | 公安部交通管理科学研究所 | Early-warning system for inspecting and judging urban road traffic safety situation |
| CN102231231A (en) * | 2011-06-16 | 2011-11-02 | 同济大学 | Area road network traffic safety situation early warning system and method thereof |
| CN102368354A (en) * | 2011-10-19 | 2012-03-07 | 北京航空航天大学 | Road security evaluation method based on floating vehicle data acquisition |
| CN104978853A (en) * | 2014-04-01 | 2015-10-14 | 中国移动通信集团公司 | Road traffic safety evaluating method and system |
| CN105608902A (en) * | 2016-03-28 | 2016-05-25 | 辽宁省交通科学研究院 | Expressway black spot identification system and method |
Non-Patent Citations (1)
| Title |
|---|
| 铁路客运枢纽公共交通安全保障能力变权评价;郭洪洋等;《重庆交通大 学学报(自然科学版)》;20160430;第35卷(第2 期);第147页第1节至第148页第2节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108182802A (en) | 2018-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Analysis of hourly crash likelihood using unbalanced panel data mixed logit model and real-time driving environmental big data | |
| CN104103194B (en) | Highway is traffic safety driving assistance method and backup system initiatively | |
| US20210334865A1 (en) | Used Car Pricing Based on Telematics Information | |
| Zolali et al. | A behavioral model of drivers’ mean speed influenced by weather conditions, road geometry, and driver characteristics using a driving simulator study | |
| Wang et al. | Causation analysis of crashes and near crashes using naturalistic driving data | |
| CN110276954A (en) | Vehicle driving behavior integration methods of marking based on BEI-DOU position system | |
| Anderson et al. | Heavy-vehicle crash rate analysis: Comparison of heterogeneity methods using Idaho crash data | |
| CN108182802B (en) | Traffic safety analysis method based on information attenuation model and driven by license plate data mining | |
| Kantor et al. | Design of algorithms for payment telematics systems evaluating driver's driving style | |
| Sanusi et al. | Cases of road traffic accident in Nigeria: a time series approach | |
| KR20130092915A (en) | Method for calculating vehicle safety driving index in safety driving index calculating system, method for calculating issurance of vehicle in safety driving index calculating system and safety driving index calculating system using the same | |
| CN106780049B (en) | Vehicle financial credit granting method and device | |
| Thapa et al. | Assessing driver behavior in work zones: A discretized duration approach to predict speeding | |
| Christoforou et al. | Integrating real-time traffic data in road safety analysis | |
| Abuaddous et al. | Identification and Ranking of Accident Black Spots in Jordan | |
| Chapelon et al. | Road safety in France: The hard path toward science-based policy | |
| Ambros et al. | Identification of road horizontal alignment inconsistencies–A pilot study from the Czech Republic | |
| Chen et al. | How expressway geometry factors contribute to accident occurrence? A binary logistic regression study | |
| Amir et al. | Examination of Differentiated Per-Lane Speed Limits on Improving Traffic Safety: Microscopic Traffic Simulation Approach | |
| AlAmin et al. | Analyzing factors associated with fatality rate of Road collisions: A case study in tennessee | |
| CN116824864B (en) | Government affair data processing system capable of flexibly configuring visual access | |
| Kriswardhana et al. | Probability of crash involvement based on riders’ behaviour | |
| Shaon | Identifying and incorporating driver behavior variables into crash prediction models | |
| WO2017221038A1 (en) | Method of determining cost of insurance based on the correlation of information from actual accidents with the historical analysis of vehicle movement | |
| Zaini et al. | Risky Behavior and Traffic Safety of Motorcycle Riders in Sungai Penuh City, Jambi Province |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200917 Address after: Room 701-1, building 1, No. 1326, Wenyi West Road, Cangqian street, Yuhang District, Hangzhou City, Zhejiang Province Applicant after: Hangzhou Yuantiao Science and Technology Co.,Ltd. Address before: Room 73, No. 34, No. 2000, Tianmu Mountain Road, Xihu District, Hangzhou, Zhejiang Applicant before: Xia Yingjie |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |