CN102722989B - Expressway microclimate traffic early warning method based on fuzzy neural network - Google Patents
Expressway microclimate traffic early warning method based on fuzzy neural network Download PDFInfo
- Publication number
- CN102722989B CN102722989B CN201210222783.9A CN201210222783A CN102722989B CN 102722989 B CN102722989 B CN 102722989B CN 201210222783 A CN201210222783 A CN 201210222783A CN 102722989 B CN102722989 B CN 102722989B
- Authority
- CN
- China
- Prior art keywords
- traffic
- layer
- microclimate
- output
- neural network
- 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.)
- Expired - Fee Related
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
Abstract
The invention discloses an expressway microclimate traffic early warning method based on a fuzzy neural network. The expressway microclimate traffic early warning method comprises the following steps of distributing traffic flow and microclimate monitoring points; defining a traffic controller of the fuzzy neural network; training the traffic controller of the fuzzy neural network; using the optimal traffic controller of the fuzzy neural network to generate traffic safety travelling parameters; and issuing traffic control information. The expressway microclimate traffic early warning method uses the fuzzy neural network and issues vehicle operating limiting-velocity value, distance limiting value and overtaking limitation and lane changing limitation measures through comprehensive detection of meteorological parameters such as precipitation, snow quantity, temperature and visibility along the line of an expressway. The method is used in the expressway to improve travelling safety under severe weather conditions.
Description
Technical field
The present invention relates to a kind of traffic safety technology, especially a kind of highway microclimate traffic prewarning method based on fuzzy neural network.
Background technology
At present, along with the increase of highway mileage open to traffic, diastrous weather highlights increasingly on the impact of expressway traffic safety.Under inclement weather conditions, the more difficult information warning that obtains in time of driver, thereby cause the major accident of hundreds of rear-end collision to happen occasionally, therefore, highway can only be closed in boisterous situation, quick early warning and the intelligent management of highway, be more and more subject to people's concern and attention.
By retrieval paper, find: Cheng Conglan, Li Xun, Zheng Zuofang, Wang Wen, Liang Xudong. Beijing road weather warning index builds and Preliminary Applications, the 27th Meteorology Society of China's year proceeding, 2010, 10. Feng Min learns, the meteorological intelligent early warning system research that detects of freeway traffic, Nanjing Information engineering Univ's PhD dissertation, 2005. long monarchs, Zou Kaiqi. the research of highway NN control system under bad weather condition, computer engineering and application, 2007, 43(4): 210-212. king flies less, pass can. expressway weather information service system. and Chinese transportation information industry, 2007(1): on the basis that the above-mentioned research of 116-119. affects the traffic capacity in analysis DIFFERENT METEOROLOGICAL CONDITIONS, for DIFFERENT METEOROLOGICAL CONDITIONS, provide weather warning information.But, directly from the angle of freeway management, do not provide traffic prewarning information.Tang Junjun, high sea otter, Zhang Weihan. the research of Supervising System Structure for Expressways in Fog Area, highway, 2005, 8. Wang Wei Asia. the monitoring of highway terrible weather and traffic control model investigation based on radio sensing network, Chang An University, 2008. these people of willow, highway security of operation control technology research under diastrous weather, Tongji University's doctorate paper, 2008. above-mentioned researchs analyze visibility and coefficient of road adhesion on the prerequisite of express way driving safety impact under, by fuzzy control theory or meteorological department, freeway management department and line driver practical experience and a test, to mist, snow, speed limit and safe spacing under rainy day gas stipulate.There is three point problem: the one,,, on the basis of comprehensive weather monitoring, vehicle driving not being carried out safely to early warning, Consideration is comprehensive not; The 2nd,, fuzzy reasoning degree of membership is directly provided by experience, and subjectivity is strong.The 3rd,, do not consider the impact of road traffic flow situation.
Chinese patent application 200910061448.3 discloses a kind of expressway weather monitoring system.And application number: 200910060562.4 disclose a kind of highway rear-end collision prevention early warning system.Chinese patent application 200710077671.8 discloses a kind of weather information prompt system for speedway.Foregoing invention all designs for Forewarning System of Freeway hardware facility, does not relate on data acquisition basis, as obtains the method for safe driving parameter and traffic prewarning.
Summary of the invention
The object of the invention is for overcoming above-mentioned the deficiencies in the prior art, a kind of highway microclimate traffic prewarning method based on fuzzy neural network is provided, the method is by the comprehensive detection of the meteorologic parameters such as highway rainfall amount along the line, snowfall, temperature, visibility, utilize the method for fuzzy neural network, issue vehicle operating speed limit, spacing limits value, overtake other vehicles restriction and the measures such as restriction of changing trains.On highway, apply the method, can, under severe weather conditions, improve travel safety.
For achieving the above object, the present invention adopts following technical proposals:
A highway microclimate traffic prewarning method based on fuzzy neural network, comprises the following steps: the laying of traffic flow and microclimate check point; Ambiguity in definition neural network traffic controller; The training of fuzzy neural network traffic controller; Utilize optimum fuzzy neural network traffic controller to generate traffic safety driving parameter; Traffic control information issue, concrete operation step is as follows:
Step 1: at interval of certain distance, some traffic flows and microclimate monitoring point are set at highway trackside, detect this road traffic delay situation and microclimate supplemental characteristic, pass through sensor Data Fusion, obtain road traffic delay and weather information, traffic flow and microclimate data acquisition are prepared for next step Fuzzy Neural-network Control;
Step 2: adopt based on Takagi-Sugeno(Gao Mu-Guan Ye) fuzzy neural network of reasoning builds highway microclimate traffic controller, telecommunication flow information and weather information that definition step 1 collects are state variable, as the input value of described controller, definition highway control mode, speed limit and safe spacing value are control variable, as the output valve of described controller;
Step 3: the meteorology, traffic flow, control measure and the implementation result historical data base thereof that adopt meteorological department and vehicle supervision department, in conjunction with expertise, build the training sample of fuzzy neural network traffic controller, highway microclimate traffic controller is trained, training error is reduced to predetermined threshold or reaches frequency of training, to obtain optimal controller;
Step 4: by the fuzzy neural network traffic controller of the telecommunication flow information of Real-time Collection and microclimate input information optimum, generation, for the Freeway Traffic Control scheme of this moment traffic flow situation and weather information, comprises Forewarning Measures, speed limit and safe spacing value;
Step 5: lay variable information plate in each traffic and upstream, microclimate monitoring point, the safe driving early warning information of traffic controller output is issued.
In described step 2, fuzzy neural network structure highway microclimate traffic controller is divided into five-layer structure:
Ground floor is input layer, and input value is the supplemental characteristic that freeway traffic flow and microclimate check point gather, and is expressed as x={x
1, x
2, x
3..., x
n, wherein, n represents the number of input parameter, n is more than or equal to 1 integer, and x
1, x
2... x
nrepresent that respectively traffic flow parameter is that speed, flow, occupation rate and microclimate parameter are visibility, temperature, humidity, air pressure, wind direction, wind speed, surface temperature;
Input parameter is divided into 5 grades according to order from small to large, is respectively { NB is negative large, and NS is negative little, Z zero, PS is just little, PB is honest }, its meaning be corresponding parameter index value be little, less, medium, greatly, larger; The system of described controller is output as control model, speed limit and safe distance; Wherein, control model is divided into three kinds, and sealing completely, region sealing and ring road control are controlled and sailed highway flow into according to traffic and meteorological condition; Speed limit and safe distance grade classification mode are identical with input parameter;
The second layer is obfuscation layer, be used for representing input quantity belong to respectively NB, NS, Z, PS, the degree of membership of PB} is
in formula, j is 1 to m
iinteger, m
ix
ifuzzy partition number, m herein
i=5; c
ijand σ
ijrepresent respectively center and the width of membership function;
The 3rd layer is regular former piece layer, and each node represents a fuzzy rule, and its effect is the former piece for mating fuzzy rule, calculates the relevance grade a of every rule
j, formula is:
Or
In formula: a
jthe relevance grade of-fuzzy rule j;
-input x
ibe under the jurisdiction of the degree of membership of j grade, j is the integer between 1 to m, and m is more than or equal to 1 integer;
Realize normalization for the 4th layer and calculate, formula is:
Layer 5 output layer, formula is
Wherein, w
rfor connection weight, r is the integer between 1 to m, and m is more than or equal to 1 integer.
The training of the microclimate traffic controller based on fuzzy neural network in described step 3, particular content is as follows:
(1) training sample chooses
Training sample is considered to be comprised of three parts: Part I, and the historical data of meteorological part, freeway traffic regulation department, comprises the effect after historical meteorologic parameter information, traffic flow parameter information, early warning information and the issue thereof taked at that time; Part II, traffic engineering domain expert's questionnaire, by arranging different meteorologies and traffic flow sight, the Forewarning Measures that expert may adopt; Part III, after this system is built up, is added into historical data base by the traffic prewarning scheme implementation effect under different meteorologies, traffic flow situation;
(2) training algorithm
The parameter that network need to be learnt is the central value c of second layer membership function in step 2
ijand width cs
ijand layer 5 network connection weight w
r, the learning algorithm of this network is selected backpropagation BP algorithm; BP algorithm is comprised of forward-propagating and two processes of error back propagation; In forward-propagating process, input message is successively processed through Hidden unit from input layer, after all hidden layers, is transmitted to output layer; In the process of successively processing at hidden layer, the neuronic state of every one deck only exerts an influence to the neuronic state of lower one deck; At output layer, reality output and desired output are compared, if the difference of actual output and desired output is no longer within tolerance interval, proceed to back-propagation process, error between actual value and desired output is returned along original connecting path, by revising each layer of neuronic connection weight, error is reduced, and then proceed to forward-propagating process, so repeatedly calculate, until error is less than setting value;
The learning algorithm of described network is selected backpropagation BP algorithm, and step is as follows:
1. establish x
kfor input vector, x
k=(x
1, x
2..., x
n), k is the integer between 1 to K, in formula: K is number of samples, and the number of n representative feature parameter; The output vector of corresponding travel pattern is y
k, initialization network weight, threshold value;
2. the each unit of the second layer is input as
In formula, x
ithe input of expression system, i.e. traffic flow parameter and meteorologic parameter value, i is the integer between 1 to n, the number of n representative feature parameter; J is 1 to m
ibetween integer, m
irepresent the fuzzy partition number of i characteristic parameter;
Transport function between ground floor and the second layer, is membership function:
Second layer unit is output as:
3. the input of the 3rd layer of each unit is the output of the corresponding each unit of the second layer, for
Output layer is output as
In formula, l is the integer between 1 to m, and m represents number of fuzzy rules;
4. the input of the 4th layer of each unit is the output of the 3rd layer of corresponding each unit, for
Output layer is output as
5. the input of the each unit of layer 5 is the output of the 4th layer of corresponding each unit, for
Output layer is output as
So far complete a forward pass process;
6. in error back propagation process, first to carry out error calculating,
For fuzzy neural network, suppose t the error function E that sample is right
tbe defined as:
In formula: y
0(t) be system desired output, y (t) is system real output value, and t is more than or equal to 1 integer, represents the label of sample.
Backpropagation BP thought is used to supervised learning, by adjusting each weighted value of network, makes error function value minimum, thereby reaches the object of revising membership function parameter and network connection weight;
7. choose at random next sample to offering network, double counting process, until network global error function
be less than predefined minimal value, i.e. a network convergence; Or study number of times be less than predefined value, network cannot be restrained; Wherein, K is learning sample number;
8. finish study.
Takagi-Sugeno(Gao Mu-Guan Ye in the present invention) fuzzy neural network of reasoning is known technology, do not repeat them here.
The invention has the beneficial effects as follows, highway microclimate traffic control method is according to the situation of the real-time meteorology of highway and traffic flow, and self-adaptation is adjusted traffic control scheme, has improved travel safety and the traffic efficiency of highway under inclement weather.There is following difference compared with traffic prewarning method under other inclement weathers:
1, control law does not need to provide in advance, has adopted historical data and expertise to model training, has improved the accuracy of control program and has abandoned the subjectivity that traffic administration person carries out traffic administration;
2, for the information of Real-time Collection, carry out the generation of traffic control scheme, and utilize at its upstream the issue of variable information plate implementation information, improved the efficiency of inclement weather traffic prewarning.
Accompanying drawing explanation
Fig. 1 microclimate detects traffic prewarning system schematic;
Fig. 2 is traffic flow and microclimate measuring points placement scheme schematic diagram;
Embodiment
Below in conjunction with accompanying drawing and example, the present invention is further described.
The structural drawing of the microclimate traffic prewarning system based on fuzzy neural network as shown in Figure 1.This system is comprised of three subsystems: subsystem is unified, traffic and microclimate data acquisition system (DAS); Subsystem two, fuzzy neural network controller; Subsystem three, traffic prewarning information issuing system.The relation of these three subsystems is as follows: the unified traffic gathering of subsystem and microclimate information are as input value, be input to subsystem two fuzzy neural network controller, through the calculating output traffic prewarning scheme of this controller, traffic prewarning scheme input subsystem three, is published to variable information plate as early warning information.
Step 1: traffic flow and microclimate detector layout
Traffic flow and microclimate measuring points placement scheme are as shown in Figure 2.In Fig. 1, module 1 represents traffic flow and microclimate monitoring modular, and it is laid in trackside according to certain intervals.Earth coil is laid in traffic flow monitoring point, to detect the data such as this section traffic flow speed of a motor vehicle, flow and occupation rate; The sensors such as microclimate detector layout visibility detecting device, Temperature Detector, surface temperature detecting device, moisture detector, rainfall detection device, freezing detecting device, gas pressure detector, sand and dust detecting device, hail detecting device, accumulated snow detecting device and sand and dust detecting device, to detect this area meteorological information.Traffic flow and microclimate data acquisition will be prepared for next step Fuzzy Neural-network Control.
Step 2: the design of fuzzy neural network traffic controller
Described fuzzy neural network traffic controller is realized traffic prewarning information issue under inclement weather, be that some historical datas, priori or traffic engineering domain expert experience are included in fuzzy rule, be convenient to obtain reasonably and traffic prewarning information meteorological and that telecommunication flow information adapts.Fuzzy neural network traffic controller is divided into five-layer structure:
Ground floor is input layer, and input value is the supplemental characteristic that freeway traffic flow and microclimate check point gather, and is expressed as x={x
1, x
2, x
3..., x
n, wherein, n represents the number of input parameter, and x
1, x
2... x
nrepresent respectively traffic flow parameter (speed, flow, occupation rate) and microclimate parameter (visibility, temperature, humidity, air pressure, wind direction, wind speed, surface temperature etc.).
Input parameter is divided into 5 grades according to order from small to large, is respectively { NB is negative large, and NS is negative little, Z zero, PS is just little, PB is honest }, its meaning be corresponding parameter index value be little, less, medium, greatly, larger.System is output as control model, speed limit and safe distance.Wherein, control model is divided into three kinds, sealing completely, region sealing and ring road control (according to traffic and meteorological condition, control and sail highway flow into); Speed limit and safe distance grade classification mode are identical with input parameter.
The network second layer is obfuscation layer, be used for representing input quantity belong to respectively NB, NS, Z, PS, the degree of membership of PB} is
j=1 in formula, 2 ..., m
i, m
ix
ifuzzy partition number, m herein
i=5.C
ijand σ
ijrepresent respectively center and the width of membership function;
The 3rd layer is regular former piece layer, and each node represents a fuzzy rule, and its effect is the former piece for mating fuzzy rule, calculates the relevance grade of every rule, and formula is:
Or
In formula: a
jthe relevance grade of-fuzzy rule j;
-input x
ibe under the jurisdiction of the degree of membership of j grade, j is the integer between 1 to m, and m is more than or equal to 1 integer;
Realize normalization for the 4th layer and calculate, formula is:
Layer 5 output layer, formula is
Wherein, w
rfor connection weight,
Step 3: the training of the microclimate traffic controller based on fuzzy neural network
(3) training sample chooses
Training sample is considered to be comprised of three parts: Part I, and the historical data of meteorological part, freeway traffic regulation department, comprises the effect after historical meteorologic parameter information, traffic flow parameter information, early warning information and the issue thereof taked at that time; Part II, traffic engineering domain expert's questionnaire, by arranging different meteorologies and traffic flow sight, the Forewarning Measures that expert may adopt; Part III, after this system is built up, is added into historical data base by the traffic prewarning scheme implementation effect under different meteorologies, traffic flow situation.
(4) training algorithm
The parameter that network need to be learnt is mainly the central value c of second layer membership function
ijand width cs
ijand layer 5 network connection weight w
r.The learning algorithm of this network is selected backpropagation BP(Back Propagation) algorithm.BP algorithm is comprised of forward-propagating and two processes of error back propagation.In forward-propagating process, input message is successively processed through Hidden unit from input layer, after all hidden layers, is transmitted to output layer.In the process of successively processing at hidden layer, the neuronic state of every one deck only exerts an influence to the neuronic state of lower one deck.At output layer, reality output and desired output are compared, if the difference of actual output and desired output is no longer within tolerance interval, proceed to back-propagation process, error between actual value and desired output is returned along original connecting path, by revising each layer of neuronic connection weight, error is reduced, and then proceed to forward-propagating process, so repeatedly calculate, until error is less than setting value.
The specific algorithm step of BP neural metwork training is as follows:
1. establish x
kfor input vector, x
k=(x
1, x
2..., x
n), k=1,2 ... K, in formula: K is number of samples, the number of n representative feature parameter; The output vector of corresponding travel pattern is y
k.Initialization network weight, threshold value and relevant parameters.
2. the each unit of the second layer is input as
In formula, m
irepresent the fuzzy partition number of i characteristic parameter.
Transport function between ground floor and the second layer, is membership function:
Second layer unit is output as:
3. the input of the 3rd layer of each unit is the output of the corresponding each unit of the second layer, for
Output layer is output as
In formula, l=1,2 ..., m.M represents number of fuzzy rules.
4. the input of the 4th layer of each unit is the output of the 3rd layer of corresponding each unit, for
Output layer is output as
5. the input of the each unit of layer 5 is the output of the 4th layer of corresponding each unit, for
Output layer is output as
So far complete a forward pass process.
6. in error back propagation process, first to carry out error calculating.
For fuzzy neural network, suppose that t the right error function of sample is defined as:
In formula: y
0(t) be system desired output, y (t) is system real output value, and backpropagation BP thought is used to supervised learning, by adjusting each weighted value of network, make error function value minimum, thereby reach the object of revising membership function parameter and network connection weight.
7. choose at random next sample to offering network, double counting process, until network global error function
(wherein, K is learning sample number) is less than predefined minimal value, i.e. a network convergence; Or study number of times be less than predefined value, network cannot be restrained.
8. finish study.
Step 4: utilize optimum fuzzy neural network traffic controller to generate traffic prewarning information
The fuzzy neural network controller that the meteorologic parameter of Real-time Collection and traffic flow parameter input are trained, generate real-time traffic prewarning information, comprise highway control mode (close completely, region is closed, ring road control etc.), speed limit and safety distance.
Step 5: early warning information issue
Utilize wireless communication technology, traffic prewarning information is sent to the variable information plate (module 2 as shown in Figure 1) of trackside, the information that variable information plate is issued should be the traffic prewarning information that its downstream controller obtains.
Although above-mentioned, by reference to the accompanying drawings the specific embodiment of the present invention is described; but not limiting the scope of the invention; one of ordinary skill in the art should be understood that; on the basis of technical scheme of the present invention, those skilled in the art do not need to pay various modifications that creative work can make or distortion still in protection scope of the present invention.
Claims (4)
1. the highway microclimate traffic prewarning method based on fuzzy neural network, is characterized in that, comprises the following steps: the laying of traffic flow and microclimate check point; Ambiguity in definition neural network traffic controller; The training of fuzzy neural network traffic controller; Utilize optimum fuzzy neural network traffic controller to generate traffic safety driving parameter; Traffic control information issue, concrete operation step is as follows:
Step 1: at interval of distance, some traffic flows and microclimate monitoring point are set at highway trackside, road traffic delay situation and microclimate supplemental characteristic in assay intervals distance, pass through sensor Data Fusion, obtain road traffic delay and weather information, traffic flow and microclimate data acquisition are prepared for next step Fuzzy Neural-network Control;
Step 2: adopt the fuzzy neural network based on Takagi-Sugeno reasoning to build highway microclimate traffic controller, telecommunication flow information and weather information that definition step 1 collects are state variable, as the input value of described controller, definition highway control model, speed limit and safe distance are control variable, as the output valve of described controller;
Step 3: the meteorology, traffic flow, control measure and the implementation result historical data base thereof that adopt meteorological department and vehicle supervision department, in conjunction with expertise, build the training sample of fuzzy neural network traffic controller, highway microclimate traffic controller is trained, training error is reduced to predetermined threshold or reaches frequency of training, to obtain optimal controller;
Step 4: by the fuzzy neural network traffic controller of the telecommunication flow information of Real-time Collection and microclimate input information optimum, generation, for the Freeway Traffic Control scheme of arithmetic for real-time traffic flow situation and weather information, comprises Forewarning Measures, speed limit and safe spacing value;
Step 5: lay variable information plate in each traffic and upstream, microclimate monitoring point, the safe driving early warning information of traffic controller output is issued.
2. the highway microclimate traffic prewarning method based on fuzzy neural network as claimed in claim 1, is characterized in that, in described step 2, fuzzy neural network structure highway microclimate traffic controller is divided into five-layer structure:
Ground floor is input layer, and input value is the supplemental characteristic that freeway traffic flow and microclimate check point gather, and is expressed as x={x
1, x
2, x
3..., x
n, wherein, n represents the number of input parameter, n is more than or equal to 1 integer, and x
1, x
2... x
nrepresent that respectively traffic flow parameter is that speed, flow, occupation rate and microclimate parameter are visibility, temperature, humidity, air pressure, wind direction, wind speed, surface temperature;
Input parameter is divided into 5 grades according to order from small to large, is respectively { NB is negative large, and NS is negative little, Z zero, PS is just little, PB is honest }, its meaning be corresponding parameter index value be little, less, medium, greatly, larger; The system of described controller is output as control model, speed limit and safe distance; Wherein, control model is divided into three kinds, and sealing completely, region sealing and ring road control are controlled and sailed highway flow into according to traffic and meteorological condition; Speed limit and safe distance grade classification mode are identical with input parameter;
The second layer is obfuscation layer, be used for representing input quantity belong to respectively NB, NS, Z, PS, the degree of membership of PB} is
in formula, j is 1 to m
iinteger, m
ix
ifuzzy partition number, m herein
i=5; c
ijand σ
ijrepresent respectively center and the width of membership function;
The 3rd layer is regular former piece layer, and each node represents a fuzzy rule, and its effect is the former piece for mating fuzzy rule, calculates the relevance grade α of every rule
j, formula is:
Or
In formula: α
jthe relevance grade of-fuzzy rule j;
-input x
ibe under the jurisdiction of the degree of membership of j grade; J is 1 to m
ibetween integer, m
ix
ifuzzy partition number, m
ifor being more than or equal to 1 integer;
Realize normalization for the 4th layer and calculate, formula is:
wherein, α
jthe relevance grade of-fuzzy rule j; J is 1 to m
ibetween integer, m
ix
ifuzzy partition number, m
ifor being more than or equal to 1 integer;
Layer 5 output layer, formula is
Wherein, w
rfor connection weight, α
rthe relevance grade of-fuzzy rule r, r is 1 to m
ibetween integer, m
ix
ifuzzy partition number, m
ifor being more than or equal to 1 integer.
3. the highway microclimate traffic prewarning method based on fuzzy neural network as claimed in claim 2, is characterized in that, the training of the microclimate traffic controller based on fuzzy neural network in described step 3, and particular content is as follows:
Choosing of training sample
Training sample is considered to be comprised of three parts: Part I, and the historical data of meteorological part, freeway traffic regulation department, comprises the effect after historical meteorologic parameter information, traffic flow parameter information, early warning information and the issue thereof taked at that time; Part II, traffic engineering domain expert's questionnaire, by arranging different meteorologies and traffic flow sight, the Forewarning Measures that expert may adopt; Part III, after this system is built up, is added into historical data base by the traffic prewarning scheme implementation effect under different meteorologies, traffic flow situation;
Training algorithm
The parameter that network need to be learnt is the central value c of second layer membership function in step 2
ijand width cs
ijand layer 5 network connection weight w
r, the learning algorithm of this network is selected backpropagation BP algorithm; BP algorithm is comprised of forward-propagating and two processes of error back propagation; In forward-propagating process, input message is successively processed through Hidden unit from input layer, after all hidden layers, is transmitted to output layer; In the process of successively processing at hidden layer, the neuronic state of every one deck only exerts an influence to the neuronic state of lower one deck; At output layer, reality output and desired output are compared, if the difference of actual output and desired output is not within tolerance interval, proceed to back-propagation process, error between actual value and desired output is returned along original connecting path, by revising each layer of neuronic connection weight, error is reduced, and then proceed to forward-propagating process, so repeatedly calculate, until error is less than setting value.
4. the highway microclimate traffic prewarning method based on fuzzy neural network as claimed in claim 3, is characterized in that, the learning algorithm of described network is selected backpropagation BP algorithm, and step is as follows:
1. establish x
kfor input vector, x
k=(x
k1, x
k2..., x
kn), k is the integer between 1 to K, in formula: K is number of samples, and the number of n representative feature parameter; The output vector of corresponding travel pattern is y
k, initialization network weight, threshold value;
2. the each unit of the second layer is input as
In formula, x
ithe input of expression system, i.e. traffic flow parameter and meteorologic parameter value; I is the integer between 1 to n, the number of n representative feature parameter; J is 1 to m
ibetween integer, m
ix
ifuzzy partition number;
Transport function between ground floor and the second layer, is membership function:
Second layer unit is output as:
3. the input of the 3rd layer of each unit is the output of the corresponding each unit of the second layer, for
Output layer is output as
In formula, l is 1 to m
ibetween integer, m
irepresent number of fuzzy rules;
4. the input of the 4th layer of each unit is the output of the 3rd layer of corresponding each unit, for
Output layer is output as
5. the input of the each unit of layer 5 is the output of the 4th layer of corresponding each unit, for
Output layer is output as
So far complete a forward pass process;
6. in error back propagation process, first to carry out error calculating,
For fuzzy neural network, suppose t the error function E that sample is right
tbe defined as:
In formula: y
0(t) be system desired output, y (t) is system real output value, and t is more than or equal to 1 integer, represents the label of sample;
Backpropagation BP thought is used to supervised learning, by adjusting each weighted value of network, makes error function value minimum, thereby reaches the object of revising membership function parameter and network connection weight;
7. choose at random next sample to offering network, double counting process, until network global error function
be less than predefined minimal value, i.e. a network convergence; Or study number of times be less than predefined value, network cannot be restrained; Wherein, K is learning sample number;
8. finish study.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210222783.9A CN102722989B (en) | 2012-06-29 | 2012-06-29 | Expressway microclimate traffic early warning method based on fuzzy neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210222783.9A CN102722989B (en) | 2012-06-29 | 2012-06-29 | Expressway microclimate traffic early warning method based on fuzzy neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102722989A CN102722989A (en) | 2012-10-10 |
CN102722989B true CN102722989B (en) | 2014-05-07 |
Family
ID=46948733
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210222783.9A Expired - Fee Related CN102722989B (en) | 2012-06-29 | 2012-06-29 | Expressway microclimate traffic early warning method based on fuzzy neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102722989B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111462493A (en) * | 2020-04-10 | 2020-07-28 | 北京工业大学 | Overtaking situation perception early warning system based on internet of vehicles |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103852810A (en) * | 2012-12-05 | 2014-06-11 | 中国北车股份有限公司 | Weather real-time detection device |
CN103646513A (en) * | 2013-11-27 | 2014-03-19 | 武汉地大信息工程股份有限公司 | Geological disaster early warning system and early warning method |
CN104112357B (en) * | 2014-07-04 | 2016-01-13 | 浙江大学 | A kind of urgent emergency planning method of city area-traffic of severe haze weather |
CN104133372B (en) * | 2014-07-09 | 2016-09-28 | 河海大学常州校区 | Room temperature control algolithm based on fuzzy neural network |
CN104318780B (en) * | 2014-10-31 | 2016-07-13 | 重庆大学 | Consider the freeway incident detection method of meteorological factor, road alignment factor |
CN105425789A (en) * | 2015-11-02 | 2016-03-23 | 合肥工业大学 | Control logic structure of unmanned mine locomotive system based on rule matching |
US10049284B2 (en) * | 2016-04-11 | 2018-08-14 | Ford Global Technologies | Vision-based rain detection using deep learning |
CN106530770A (en) * | 2016-12-01 | 2017-03-22 | 清华大学 | Agglomerate fog road section driving safety intelligent detection and early warning method and system |
CN108305479B (en) * | 2018-03-13 | 2019-12-03 | 辽宁工业大学 | A kind of traffic induction method to turn left for traffic intersection |
CN108710359B (en) * | 2018-05-04 | 2021-03-02 | 江苏科技大学 | Remote fault diagnosis system for ship power device and determination method for fuzzification function |
CN108830032A (en) * | 2018-05-17 | 2018-11-16 | 南京邮电大学 | A kind of unmanned plane weather warning method neural network based |
CN108898697A (en) * | 2018-07-25 | 2018-11-27 | 广东工业大学 | A kind of road surface characteristic acquisition methods and relevant apparatus |
CN110516353B (en) * | 2019-08-27 | 2024-03-26 | 浙江科技学院 | Method for rapidly identifying design defects of expressway curve in mountain area |
CN110910641A (en) * | 2019-12-02 | 2020-03-24 | 北京航空航天大学 | Neural network-based real-time risk assessment method for highway traffic operation |
CN111739289B (en) * | 2020-06-02 | 2024-02-20 | 腾讯科技(深圳)有限公司 | Method and device for processing vehicle early warning information |
CN111835715B (en) * | 2020-06-03 | 2021-06-04 | 北京邮电大学 | Method and device for determining safety value of virtual network function |
CN113064220A (en) * | 2021-06-03 | 2021-07-02 | 四川九通智路科技有限公司 | Visibility measuring system and measuring method based on nonlinear autoregressive neural network |
CN113506441B (en) * | 2021-09-08 | 2021-11-30 | 苏州博宇鑫交通科技有限公司 | Municipal bridge traffic early warning control method |
CN113781779A (en) * | 2021-09-09 | 2021-12-10 | 济南金宇公路产业发展有限公司 | 5G communication-based highway weather early warning method, equipment and medium |
CN113888850B (en) * | 2021-12-03 | 2022-03-04 | 南京坤拓土木工程科技有限公司 | Sand-dust meteorological municipal road congestion situation multi-factor early warning method |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0436900A (en) * | 1990-05-31 | 1992-02-06 | Omron Corp | Road information presenting device |
CN101149819A (en) * | 2007-10-31 | 2008-03-26 | 山东省科学院海洋仪器仪表研究所 | Meteorological element real time data singular value elimination method |
CN101261765A (en) * | 2007-03-06 | 2008-09-10 | 张小平 | Weather information prompt system for speedway |
CN101609606A (en) * | 2009-07-10 | 2009-12-23 | 清华大学 | A kind of road surface wet and slippery state identification method |
CN101739828A (en) * | 2009-11-18 | 2010-06-16 | 大连理工大学 | Urban traffic area jamming judgment method by combining road traffic and weather state |
CN101746269A (en) * | 2010-01-08 | 2010-06-23 | 东南大学 | Fatigue driving fusion detection method based on soft computing |
CN101859492A (en) * | 2009-04-07 | 2010-10-13 | 武汉大学 | Expressway weather monitoring system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7689348B2 (en) * | 2006-04-18 | 2010-03-30 | International Business Machines Corporation | Intelligent redirection of vehicular traffic due to congestion and real-time performance metrics |
-
2012
- 2012-06-29 CN CN201210222783.9A patent/CN102722989B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0436900A (en) * | 1990-05-31 | 1992-02-06 | Omron Corp | Road information presenting device |
CN101261765A (en) * | 2007-03-06 | 2008-09-10 | 张小平 | Weather information prompt system for speedway |
CN101149819A (en) * | 2007-10-31 | 2008-03-26 | 山东省科学院海洋仪器仪表研究所 | Meteorological element real time data singular value elimination method |
CN101859492A (en) * | 2009-04-07 | 2010-10-13 | 武汉大学 | Expressway weather monitoring system |
CN101609606A (en) * | 2009-07-10 | 2009-12-23 | 清华大学 | A kind of road surface wet and slippery state identification method |
CN101739828A (en) * | 2009-11-18 | 2010-06-16 | 大连理工大学 | Urban traffic area jamming judgment method by combining road traffic and weather state |
CN101746269A (en) * | 2010-01-08 | 2010-06-23 | 东南大学 | Fatigue driving fusion detection method based on soft computing |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111462493A (en) * | 2020-04-10 | 2020-07-28 | 北京工业大学 | Overtaking situation perception early warning system based on internet of vehicles |
CN111462493B (en) * | 2020-04-10 | 2021-05-28 | 北京工业大学 | Overtaking situation perception early warning system based on internet of vehicles |
Also Published As
Publication number | Publication date |
---|---|
CN102722989A (en) | 2012-10-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102722989B (en) | Expressway microclimate traffic early warning method based on fuzzy neural network | |
CN104200687B (en) | A kind of driver's speed control behavior monitoring device and monitoring method | |
Yang et al. | Eco-driving system for connected automated vehicles: Multi-objective trajectory optimization | |
CN103839409B (en) | Based on the traffic flow modes method of discrimination of multibreak facial vision sensing cluster analysis | |
CN102136190B (en) | Dispatching management system and method for event emergency response of urban bus passenger transport | |
CN103971523B (en) | A kind of mountain road traffic safety dynamic early-warning system | |
CN102592447B (en) | Method for judging road traffic state of regional road network based on fuzzy c means (FCM) | |
CN103164962B (en) | Mountain road sharp turn section real-time vehicle speed early warning method | |
CN102509454B (en) | Road state merging method based on floating car data (FCD) and earth magnetism detector | |
CN105912862B (en) | A kind of exhaust emissions quantity measuring method and air pollution analysis method and apparatus | |
CN104408925A (en) | Array radar based intersection running state evaluation method | |
CN106529608A (en) | Comprehensive stationing system for exhaust gas remote-measuring equipment of motor vehicle | |
CN106846806A (en) | Urban highway traffic method for detecting abnormality based on Isolation Forest | |
CN102201021B (en) | Expressway aided design system | |
CN100466010C (en) | Different species traffic information real time integrating method | |
CN115081508B (en) | Traffic running risk parallel simulation system based on traffic digital twin | |
CN115206103B (en) | Variable speed limit control system based on parallel simulation system | |
CN104575050A (en) | Express way ramp intelligent inducing method and device based on floating vehicles | |
CN109389845B (en) | Multi-factor integrated dynamic speed control system for expressway | |
CN106530763A (en) | Supersaturation traffic adaptive signal control method with coupling of inlet flow and outlet flow | |
CN110889444B (en) | Driving track feature classification method based on convolutional neural network | |
Ma et al. | Active traffic management strategies for expressways based on crash risk prediction of moving vehicle groups | |
Chen et al. | Adaptive ramp metering control for urban freeway using large-scale data | |
Alam et al. | Intellegent traffic light control system for isolated intersection using fuzzy logic | |
CN107066501A (en) | A kind of telemetering motor vehicle tail equipment points distributing method based on road similitude |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140507 Termination date: 20160629 |