CN111476463A - Intelligent detection system for highway meteorological parameters and road surface slippery - Google Patents

Intelligent detection system for highway meteorological parameters and road surface slippery Download PDF

Info

Publication number
CN111476463A
CN111476463A CN202010201762.3A CN202010201762A CN111476463A CN 111476463 A CN111476463 A CN 111476463A CN 202010201762 A CN202010201762 A CN 202010201762A CN 111476463 A CN111476463 A CN 111476463A
Authority
CN
China
Prior art keywords
temperature
neural network
detection
road surface
time
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.)
Withdrawn
Application number
CN202010201762.3A
Other languages
Chinese (zh)
Inventor
陈亚娟
陈雪松
马晨雷
马从国
丁晓红
张月红
葛红
张利兵
周恒瑞
钟洪青
丁百湛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huaiyin Institute of Technology
Original Assignee
Huaiyin Institute of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huaiyin Institute of Technology filed Critical Huaiyin Institute of Technology
Priority to CN202010201762.3A priority Critical patent/CN111476463A/en
Publication of CN111476463A publication Critical patent/CN111476463A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks

Abstract

The invention discloses an intelligent detection system for highway meteorological parameters and road surface slippery, which is characterized in that the detection system consists of a highway meteorological environment parameter acquisition platform based on a CAN bus and a road surface slippery degree grade classification system, the highway meteorological environment parameter acquisition platform based on the CAN bus realizes the detection and monitoring of highway meteorological environment factor parameters, and the road surface slippery degree grade classification system consists of a temperature detection module, a rain and snow detection module, an icing detection module and an interval number least square support vector machine L S-SVM road surface slippery classifier.

Description

Intelligent detection system for highway meteorological parameters and road surface slippery
Technical Field
The invention relates to the technical field of highway meteorological parameter automatic equipment, in particular to an intelligent detection system for highway meteorological parameters and road surface slippery conditions.
Background
The highway plays a significant role in the transportation industry due to the high transportation speed and high efficiency. However, the road safety operation is influenced by the increasing environment of the disastrous weather such as low rainstorm, high temperature, ice and snow and the like and the derived disasters, and some serious traffic accidents are often caused. Therefore, the influence of the disastrous weather on the high-speed traffic accidents is analyzed, and the improvement of the traffic weather service quality has important significance for guaranteeing the safety, the efficiency and the economic benefit of the highway transportation system. Extreme weather caused by climate change brings great negative influence to the traffic industry, and in order to cope with the adverse influence of disastrous weather on traffic, many scholars develop research on the influence of disastrous weather such as heavy rain, heavy snow, freezing, high temperature, haze and the like on road traffic safety, and correspondingly provide forecast early warning indexes, methods and countermeasures for road safety operation under different disastrous weather. The intelligent detection system for the road meteorological parameters and the road surface slippery is invented.
Disclosure of Invention
The invention provides an intelligent detection system for highway meteorological parameters and road surface slippery, which effectively solves the problem that the conventional detection system for highway meteorological environment parameters can not detect the highway meteorological environment parameters and the road surface slippery according to the characteristics of nonlinearity, large lag, complex highway meteorological environment change and the like of the highway meteorological environment parameter change, so that the prediction of the highway safety performance is greatly influenced.
The invention is realized by the following technical scheme:
an intelligent detection system for road meteorological parameters and road surface slippery conditions is characterized in that the detection system comprises a road meteorological environment parameter acquisition platform based on a CAN bus and a road surface slippery condition grade classification system, the road meteorological environment parameter acquisition platform based on the CAN bus realizes detection and monitoring of road meteorological environment factor parameters, the road surface slippery condition grade classification system comprises a temperature detection module, a rain and snow detection module, an icing detection module and an interval number least square support vector machine L S-SVM road surface slippery condition classifier, the outputs of the temperature detection module, the rain and snow detection module and the icing detection module are used as the inputs of an interval number least square support vector machine L S-SVM road surface slippery condition classifier, the output of the interval number least square support vector machine L S-SVM road surface slippery condition classifier is an interval number representing the detected road surface slippery condition grade, and the road surface slippery condition grade classification system realizes detection and classification of the road surface slippery condition grade.
The invention further adopts the technical improvement scheme that:
the temperature detection module consists of a plurality of temperature interval number neural network models, a highway multipoint temperature fusion model and a temperature interval number prediction model, the output of all detection point temperature sensors is used as the input of the corresponding temperature interval number neural network models, the output of the temperature interval number neural network models is used as the input of the highway multipoint temperature fusion model, the output of the highway multipoint temperature fusion model is used as the input of the temperature interval number prediction model, and the output of the temperature interval number prediction model is used as the input of an interval number least square support vector machine L S-SVM road surface slippery moisture classifier;
the section number least square support vector machine L S-SVM road surface slippery wet classifier is a least square support vector machine L S-SVM with the input of 3 sections and the output of 1 section number, the output of the temperature detection module, the rain and snow detection module and the icing detection module is the input of a section number least square support vector machine L S-SVM road surface slippery wet classifier, the output of the section number least square support vector machine L S-SVM road surface slippery wet classifier is the section number representing the magnitude of the slippery wet degree grade of a detected road, according to engineering practice of slippery wet road surface on road driving and road surface grade and surface layer type (GB/T-920), the section number support vector machine L S-SVM road surface slippery wet classifier, a corresponding relation table of 5 slippery wet degree grades and 5 sections of the road surface is constructed, the 5 slippery wet degree grades of the road surface are respectively in a dry state, slippery wet degree signs, danger and extraordinary danger are compared, the section number least square support vector machine L S-SVM road surface slippery wet classifier is used for calculating the maximum degree of the slippery wet degree of the road surface and the road surface slippery degree corresponding to the section similar to the maximum wet degree of the road surface slippery degree of the road surface.
The invention further adopts the technical improvement scheme that:
the temperature interval number neural network model consists of an RR time recurrent neural network, an interval number DRNN neural network model and 3 beat delay lines TD L, the temperature interval number neural network model converts a period of time highway temperature sensor sensing measured highway temperature values into dynamic interval numerical values of highway temperature, the output of each detection point temperature sensor is the input of the corresponding RR time recurrent neural network, the output of each RR time recurrent neural network is used as the input of the corresponding beat delay line TD L, the output of 3 beat delay lines TD L is used as the input of the interval number DRNN neural network model, the output of the DRNN neural network model is the interval number formed by the output of the temperature interval number neural network model and the upper and lower limit values representing the highway temperature in a period of time, and the upper and lower limit values of the interval number of the highway temperature are respectively used as the inputs of the corresponding 2 beat delay lines TD L.
The invention further adopts the technical improvement scheme that:
the road multipoint temperature fusion model forms a time sequence temperature interval number array according to the output of the temperature interval number neural network model of all the detection points in a period of time, determines the positive and negative ideal values of the time sequence temperature interval number array, and respectively calculates the distance and similarity between the time sequence temperature interval value of each detection point and the positive and negative ideal values of the time sequence temperature interval number array; and the quotient obtained by dividing the negative ideal value distance of the time-series temperature interval numerical value of each detection point by the sum of the negative ideal value distance of the time-series temperature interval numerical value of each detection point and the positive ideal value distance of the time-series temperature interval numerical value of each detection point is the distance relative closeness of the time-series temperature interval numerical value of each detection point, and the quotient obtained by dividing the distance relative closeness of the time-series temperature interval numerical value of each detection point by the sum of the distance relative closeness of the time-series temperature interval numerical values of all the detection points is the distance fusion weight of the time-series temperature interval numerical value of each detection point.
The invention further adopts the technical improvement scheme that:
the temperature interval number prediction model consists of 2 beat delay lines TD L, 2 ANFIS neural network prediction models, 2L STM neural network residual prediction models and an interval number Elman neural network fusion model, the upper limit value and the lower limit value of the output interval number of the highway multipoint temperature fusion model are respectively input into 2 corresponding beat delay lines TD L of the temperature interval number prediction model, the output of 2 beat delay lines TD L are respectively input into 2 corresponding upper limit value and lower limit value ANFIS neural network prediction models, the difference between the upper limit value and the lower limit value of the output interval number of the highway multipoint temperature fusion model and the output of the 2 corresponding upper limit value and lower limit value ANFIS neural network prediction models is respectively input into the corresponding 2 upper limit value and lower limit value L STM neural network residual prediction models, the output of the 2 upper limit value ANFIS neural network prediction models and the output of the 2 upper limit value and lower limit value L STM neural network residual prediction models are output as the input interval number Elman neural network fusion model, and the output of the temperature interval number Elman neural network fusion model is the predicted value of the output interval number ElSTM and the temperature number Elman neural network prediction model.
The invention further adopts the technical improvement scheme that:
the quotient obtained by dividing the positive ideal value similarity of the time-series temperature interval numerical value of each detection point in a period of time by the sum of the negative ideal value similarity of the time-series temperature interval numerical value of each detection point and the positive ideal value similarity of the time-series temperature interval numerical value of each detection point is the relative closeness of the time-series temperature interval numerical values of each detection point, and the quotient obtained by dividing the relative closeness of the similarity of the time-series temperature interval numerical values of each detection point by the sum of the relative closeness of the similarity of the time-series temperature interval numerical values of all the detection points is the similarity fusion weight of the time-series temperature interval numerical values of each detection point;
the ratio of the distance fusion weight of the time sequence temperature interval numerical value of each detection point of the highway in a period of time to the root mean square of the similarity fusion weight product to the distance fusion weight of the time sequence temperature interval numerical values of all the temperature detection points and the root mean square sum of the similarity fusion weight product is the root mean square combination weight of the time sequence temperature interval numerical value fusion of the detection points; the distance fusion weight and the similarity fusion weight of the time sequence temperature interval numerical value of each detection point in a period of time are linearly combined into the linear combination weight of the time sequence temperature interval numerical value of the highway temperature detection point, the root mean square combination weight and the linear combination weight of the time sequence temperature interval numerical value fusion of each highway temperature detection point form the time sequence temperature interval numerical value fusion weight of the detection point, and the sum of the time sequence temperature interval numerical value of each detection point sensor and the time sequence temperature interval fusion weight of the detection point sensor is the fusion value of the time sequence interval numerical values of all the temperature detection points of the highway.
The invention further adopts the technical improvement scheme that:
the CAN bus-based highway meteorological environment parameter acquisition platform consists of a detection node, a control node and an on-site monitoring terminal, and communication among the detection node, the control node and the on-site monitoring terminal is realized through the CAN bus; the detection nodes respectively consist of a sensor group module, a single chip microcomputer and a communication interface, wherein the sensor group module is responsible for detecting the temperature, rain, snow, ice and visibility highway meteorological environment parameters of the highway meteorological environment, and the single chip microcomputer controls the sampling interval and sends the sampling interval to the field monitoring end through the communication module; the control node controls the adjusting equipment of the highway meteorological environment parameters; the field monitoring end consists of an industrial control computer and an RS232/CAN communication module, and is used for managing meteorological environment parameters of the detection nodes for detecting the highway and constructing a road surface slippery degree grade classification system.
The invention further adopts the technical improvement scheme that:
the rain and snow detection module consists of a plurality of rain and snow interval number neural network models, a multi-point rain and snow fusion model and a rain and snow interval number prediction model; the icing detection module consists of a plurality of icing interval number neural network models, a road multi-point icing fusion model and an icing interval number prediction model; the functional structures of the rain and snow detection module, the icing detection module and the temperature detection module have similar characteristics.
Compared with the prior art, the invention has the following obvious advantages:
the invention relates to a method for measuring road temperature, rain, snow and icing parameters, which aims at the uncertainty and randomness of the problems of sensor precision error, interference, abnormal measured road temperature, rain, snow and icing and the like in the measurement process of the road temperature, rain, snow and icing parameters.
The RNN time recursive neural network is a neural network used for processing time series data of highway temperature, rain, snow and ice. In the network, the loop structure will keep the state value of the hidden neuron at the current time and input it into the hidden layer neuron at the next time as a part of the input signal of the next loop input. The input signal of RNN adopts the time sequence input of road temperature, rain, snow and ice, each layer shares the network weight and bias when inputting one step, thereby greatly reducing the parameters needing to be learned in the network and reducing the complexity of the network.
The RNN time recursive neural network fully utilizes the correlation among time series data, is a neural network with a directional cycle structure added in a hidden layer, can well process the problems of road temperature, rain, snow and icing data based on the time series by a special structure, shows stronger ability of learning essential characteristics of a road temperature, rain, snow and icing data set by representing and inputting the distributed representation of the road temperature, rain, snow and icing data, realizes the approximation of a complex function, better describes rich intrinsic information of the road temperature, rain, snow and icing data, has stronger generalization ability, and improves the accuracy and reliability of calculating the road temperature, rain, snow and icing size.
The RNN time recursive neural network is a neural network introducing a time sequence concept, has a feedback mechanism, and is widely applied to time sequence data modeling. The RNN may store the learned information in the network, enabling the model to learn the dependency of the current time on past information. Given an input sequence, the hidden layer state ht of the RNN time recurrent neural network at any time t is based on the road temperature, rain, snow and icing input x at the current timetAnd hidden layer state h at past timet－1The state of the hidden layer at each moment can be transmitted to the next moment by the RNN time recursive neural network; and finally, the RNN time recursive neural network maps the road temperature, the rain and the snow and the ice for a period of time through the output layer to obtain the output quantity of the road temperature, the rain and the snow and the ice.
The road temperature, the rain, the snow and the ice have complex nonlinear characteristics, the road temperature, the rain, the snow and the ice change greatly under different working conditions, an accurate mathematical model is difficult to establish, the road temperature, the rain, the snow and the ice predicted value can be accurately identified by using the interval number temperature, the rain, the snow and the ice prediction module, and the mathematical model has good nonlinear approximation capability, 2 beat-to-beat delay lines TD L, 2 ANFIS neural network prediction models, 2L STM neural network residual error prediction models and an interval number Elman neural network fusion model are used in the mathematical model, the ANFIS neural network prediction model is fully applied in the mathematical model and has the reasoning function of a fuzzy reasoning system, the training and learning function of a neural network is also used, the advantages of the ANFIS neural network prediction model and the neural network fusion function of the interval number Elman neural network fusion model are combined, the characteristics of a pure neural network black box are overcome, certain transparency is achieved, and the accuracy of temperature, rain and ice prediction is improved by using the L STM neural network residual error prediction model.
Sixth, the scientificity and reliability of classification of road surface slippery wet degree grade by road temperature, rain, snow and ice are improved by the interval number least square support vector machine L S-SVM road surface slippery wet classifier of the invention, the interval number least square support vector machine L S-SVM road surface slippery wet classifier of the patent quantifies the dynamic degree of the road surface safety influence on the road stroke safety according to the engineering practice experience of road temperature, rain, snow and ice on the stroke and the road surface grade and surface layer type (GB/T920-.
Drawings
FIG. 1 is a road meteorological environment parameter acquisition platform based on a CAN bus;
FIG. 2 is a schematic diagram of a road surface slippery condition grade classification system according to the present invention;
FIG. 3 is a functional diagram of a detection node according to the present invention;
FIG. 4 is a functional diagram of a control node according to the present invention;
FIG. 5 is a functional diagram of the site monitoring software of the present invention;
FIG. 6 is a temperature interval number neural network model according to the present invention.
Detailed Description
The technical scheme of the invention is further described by combining the attached drawings 1-6:
1. design of overall system function
The invention realizes the detection of the highway meteorological environment factor parameters and the fusion and prediction of the highway meteorological environment parameters, and the system consists of a highway meteorological environment parameter acquisition platform based on a CAN bus and a road surface slippery degree grade classification system 2. The highway meteorological environment parameter acquisition platform based on the CAN bus comprises a detection node 1 for highway meteorological environment parameters and a control node 2 for adjusting the highway meteorological environment parameters, and a measurement and control network is constructed in a CAN bus mode to realize the field communication among the detection node 1, the control node 2 and a field monitoring terminal 3; the detection node 1 sends the detected road meteorological environment parameters to the field monitoring terminal 3 and carries out primary processing on the sensor data; the field monitoring terminal 3 transmits control information to the detection node 1 and the control node 2. The whole system structure is shown in figure 1.
2. Design of detection node
The invention adopts the detection node 1 based on the CAN bus as a road meteorological environment parameter sensing terminal, and the mutual information interaction between the detection node 1 and the control node 2 and the on-site monitoring terminal 3 is realized in a CAN bus mode. The detection node 1 comprises a sensor for collecting highway meteorological environment temperature, rain, snow, icing and visibility parameters, a corresponding signal conditioning circuit and a C8051F040 microprocessor; the software of the detection node mainly realizes field bus communication and acquisition and pretreatment of highway meteorological environment parameters. The software is designed by adopting a C language program, so that the compatibility degree is high, the working efficiency of software design and development is greatly improved, and the reliability, readability and transportability of program codes are enhanced. The structure of the detection node is shown in fig. 3.
3. Control node
The control node 2 is provided with a 4-channel D/A conversion circuit for regulating output quantity of temperature, icing, rain and snow and visibility, a C8051F040 microprocessor and a CAN bus communication module interface in an output channel, so as to realize regulation of control equipment influencing the highway meteorological environment, and the control node is shown in figure 4.
4. Site monitoring terminal software
The on-site monitoring terminal 3 is an industrial control computer, the on-site monitoring terminal 3 mainly realizes the collection of the road meteorological environment parameters, the multi-point detection parameter fusion and the prediction of the road meteorological environment parameters, and realizes the information interaction with the detection node 1 and the control node 2, and the on-site monitoring terminal 3 mainly has the functions of communication parameter setting, data analysis and data management, and the multi-point parameter fusion and prediction of the road meteorological environment. The management software selects Microsoft Visual + +6.0 as a development tool, an Mscomm communication control of a system is called to design a communication program, the functions of the field monitoring end software are shown in figure 5, a rain and snow interval number neural network model of a rain and snow detection module, a multi-point road rain and snow fusion model and a rain and snow interval number prediction model of a road and an icing interval number neural network model of an icing detection module, a multi-point road icing fusion model and an icing interval number prediction model of an icing detection module refer to the same design method of the temperature interval number neural network model, the multi-point road temperature fusion model and the corresponding model of the temperature interval number prediction model in the temperature detection module; the road surface slippery degree grade classification system is designed as follows:
(1) temperature interval number neural network model design
The temperature interval number neural network model consists of an RR time recurrent neural network, an interval number DRNN neural network model and 3 beat Delay lines TD L (Tapped Delay L ine), the interval number neural network model converts the temperature value of a measured road sensed by a road temperature sensor for a period of time into a dynamic interval numerical value of the road temperature, and each detection point temperature sensorThe output of the DRNN is the input of the corresponding RR time recurrent neural network, the output of the RR time recurrent neural network model is the input of the corresponding beat delay line TD L, the output of the 3 beat delay lines TD L is the input of the interval number DRNN neural network model, the output of the DRNN neural network is the interval number formed by the upper limit value and the lower limit value of the road temperature in a period of time, the upper limit value and the lower limit value of the road temperature interval number are respectively used as the input of the corresponding 2 beat delay lines TD L, and the output of the DRNN neural network is u1(k) And u2(k)，u1(k) And u2(k) Respectively as inputs to corresponding beat delay lines TD L1(k) And u2(k) Respectively representing the upper limit value and the lower limit value of the output of the numerical neural network model of the temperature interval of the detection point, and forming the numerical value of the output interval of the detected temperature of the road temperature sensor in a period of time as [ u [ ]2,u1]The neural network model identification structure of temperature interval number is shown in FIG. 6, where X (k-l), …, X (k-m) are historical data output by RR time recurrent neural network, U1(k-1)，…，U1(k-d) historical data of the upper limit value of the output value of the neural network model for the temperature interval number, U2(k-1)，…，U2(k-d) historical data of the lower limit value of the numerical neural network model output value in the temperature interval, u1(k) And u2(k) The output value of the DRNN neural network represents the output of the temperature interval numerical neural network model, k represents the current time, and m and d represent the lag points of X and U respectively. The temperature interval number neural network model can be described as:
U(k)＝[u2(k),u1(k)]＝F[X(k),X(k-1),…,X(k-m)；u1(k),…,u1(k-d)；u2(k),…,u2(k-d)](1)
① RNN time recursive neural network design
The RNN time recursive neural network can process the sequence information of the road temperature, uses the output of the previous state of the road temperature as a part of the input of the predicted next temperature, and has the function of 'memorizing' the road temperature in a general sense. The RNN temporal recurrent neural network may retain the previous sequenceAnd the road temperature is used as output, and the road temperature input of the next sequence and the reserved previous sequence are jointly calculated to obtain the road temperature output of the next sequence. x is the number oftIs the input at time t, stRepresenting the state of a memory unit of the network at time t, stState s by previous stept-1And input x at the current timetJointly calculating to obtain:
st＝f(Uxt+Wst-1) (2)
the stimulus function f is a non-linear function tanh in the RNN neural network, usually the first hidden state st-1The value of (c) will be initialized with 0, but actually initializing with a minimum value will cause the gradient to drop faster. otIs the output at time t, typically a probability vector calculated by a normalized exponential function:
ot＝softmax(Vst) (3)
②, interval number DRNN neural network design
The interval number DRNN neural network is a dynamic regression neural network with feedback and the ability of adapting to time-varying characteristics, the network can more directly and vividly reflect the dynamic change performance of the highway temperature and can more accurately predict the interval value of the highway temperature, and the 3-layer network structure of the DRNN neural network is an input layer, and a hidden layer of the DRNN neural network is a regression layer and an output layer. In the interval number DRNN neural network model of the present invention, let I ═ I1(t),I2(t),…,In(t)]Inputting a vector for the network, wherein Ii(t) is the input of the ith neuron of the DRNN neural network input layer at the t moment, and the output of the jth neuron of the regression layer is Xj(t)，Sj(t) is the input sum of the jth regression neuron, f (·) is a function of S, and O (t) is the output of the interval number DRNN neural network. The output of the DRNN neural network output layer is:
Figure BDA0002419625480000103
the output of the interval DRNN neural network is the interval number of the road temperature, and the output of the RNN time recurrent neural network in a period of time is used as the outputThe input of the DRNN neural network, the output of the DRNN neural network is the interval number of the road temperature, the output interval number of the detection point temperature sensor for detecting the road temperature in a period of time is [ u [ u ] ]2,u1]。
(2) Design of road multi-point temperature fusion model
①, constructing time series temperature interval numerical value array of temperature sensor
Interval values of the temperature sensor with a plurality of temperature detection points in a period of time form a time series temperature interval value array, nm interval values with n detection points and m moments form a time series temperature interval value array with n rows and m columns, and the time series temperature interval value of the ith detection point is set as Aij(t),Aij(t+1),…,Aij(m), the time series temperature interval number array of all the detection points is as follows:
Figure BDA0002419625480000101
②, calculating the distance fusion weight of the positive and negative ideal values of the time series temperature interval value
The average value of the temperature interval values of all the detection points at the same moment forms a positive ideal value of the time series temperature interval number array, and the positive ideal value of the time series temperature interval values is as follows:
Figure BDA0002419625480000102
the temperature interval value of the detection point at the same moment and the interval value with the maximum distance between the positive ideal value form a negative ideal value of the time series temperature interval number array, and the negative ideal value of the time series temperature interval value is as follows:
Figure BDA0002419625480000111
the distance between the time series temperature interval value of each detection point and the positive ideal value of the time series temperature interval number array is as follows:
Figure BDA0002419625480000112
the distance between the time sequence temperature interval value of each detection point and the negative ideal value of the time sequence temperature interval number array is as follows:
Figure BDA0002419625480000113
the distance relative closeness of the time series temperature interval numerical value of each detection point is obtained by dividing the negative ideal value distance of the time series temperature interval numerical value of each detection point by the sum of the negative ideal value distance of the time series temperature interval numerical value of each detection point and the positive ideal value distance of the time series temperature interval numerical value of each detection point, wherein the quotient is:
Figure BDA0002419625480000114
it can be known through the formula (10) calculation that the greater the relative closeness of the time series temperature interval numerical value of each detection point to the positive and negative ideal values of the time series temperature interval numerical array, the closer the time series temperature interval numerical value of the detection point is to the positive ideal value, otherwise, the farther the time series temperature interval numerical value of the detection point is from the positive ideal value, and according to the principle, the distance fusion weight of the time series temperature interval numerical value of each detection point obtained by dividing the relative closeness of the distance of the time series temperature interval numerical value of each detection point by the sum of the relative closeness of the distances of the time series temperature interval numerical values of all the detection points is determined as the distance fusion weight of the time series temperature interval numerical value of each detection point:
Figure BDA0002419625480000115
③, calculating the fusion weight of the positive and negative ideal value similarity of the temperature interval value
Because the temperature of the detection environment is interfered by a plurality of factors, the temperature of the ith sensor of the detection environment is continuously measured at different momentsA value of Ai1(t),Ai2(t+1)…Aim(m) defining the similarity between the interval value of the continuous measured temperature of the ith temperature detection point in a period of time and the positive ideal value as
Figure BDA0002419625480000121
The interval number similarity is calculated according to the following formula
Figure BDA0002419625480000122
The expression of (A) is as follows:
Figure BDA0002419625480000123
wherein α > 0 is a support coefficient, and the recognition rate of the similarity is improved by properly adjusting the support coefficient;
Figure BDA0002419625480000124
the distance between the temperature values of the measured environment detected by the two sensors at the same moment; if the two values are equal, the similarity of the two interval numbers is 1.
Defining the similarity between the interval value of the continuous measured temperature of the ith temperature detection point in a period of time and the negative ideal value as
Figure BDA0002419625480000125
The interval number similarity is calculated according to the following formula
Figure BDA0002419625480000126
The expression of (A) is as follows:
Figure BDA0002419625480000127
the similarity of the time-series temperature interval numerical values of each detection point obtained by dividing the positive ideal value similarity of the time-series temperature interval numerical values of each detection point by the sum of the positive ideal value similarity of the time-series temperature interval numerical values of each detection point and the negative ideal value similarity of the time-series temperature interval numerical values of each detection point is defined as the relative closeness of the time-series temperature interval numerical values of each detection point:
Figure BDA0002419625480000128
it can be known through formula (14) calculation that the greater the relative closeness of the time-series temperature interval numerical value of each detection point to the positive and negative ideal values of the time-series temperature interval numerical array, the more similar the shape of the time-series temperature interval numerical value of the detection point to the positive ideal value, otherwise, the greater the shape difference between the time-series temperature interval numerical value of the detection point and the positive ideal value, and according to this principle, the similarity fusion weight of the time-series temperature interval numerical value of each detection point obtained by dividing the relative closeness of the similarity of the time-series temperature interval numerical values of all the detection points by the sum of the relative closeness of the similarity of the time-series temperature interval numerical values of all the detection points is determined as follows:
Figure BDA0002419625480000131
④ calculating the fusion value of the temperature interval values of all the detection points
Distance fusion weight α obtained by determining distance and similarity between time series temperature interval value and positive and negative ideal values of different detection pointsiAnd similarity fusion weight βiCalculating the root mean square combination weight λiIs obviously lambdaiAnd αi、βiThe sum should be as close as possible, according to the principle of minimum relative entropy:
Figure BDA0002419625480000132
solving the optimization problem by a Lagrange multiplier method to obtain:
Figure BDA0002419625480000133
the number of time-series temperature intervals per detection point can be known from the formula (17)The ratio of the root mean square sum of the support fusion weight and the similarity fusion weight product of the root mean square of the distance fusion weight and the similarity fusion weight product of the values to the time series temperature interval values of all the detection points is the root mean square combination weight of the time series temperature interval value fusion of the detection points, and the distance fusion weight α obtained by determining the distance and the similarity between the time series temperature interval values of different detection points and the positive and negative ideal valuesiAnd similarity fusion weight βiLinear combination is carried out to obtain the linear combination weight theta of the time sequence temperature interval numerical value of the detection pointiThe formula is as follows:
θi＝ααi+ββi(18)
obtaining interval number fusion weight w according to formula (17) and formula (18)i：
wi＝[min(θi,λi),max(θi,λi)](19)
From the formula (19), the root mean square combination weight of the time-series temperature interval value fusion of each detection point and the linear combination weight constitute the interval number fusion weight of the time-series temperature interval value fusion of the detection point.
⑤, calculating the temperature interval number fusion value of all the detection points
The time sequence temperature interval number fusion value of all the detection points of the highway, which is obtained by adding the product of the time sequence temperature interval number value of each detection point and the interval number fusion weight of the time sequence interval number of the detection point, is as follows:
Figure BDA0002419625480000141
(3) temperature interval number prediction model design
The temperature interval number prediction model consists of 2 beat delay lines TD L, 2 ANFIS neural network prediction models, 2L STM neural network residual prediction models and an interval number Elman neural network fusion model, the upper limit value and the lower limit value of the output interval number of the highway multipoint temperature fusion model are respectively the input of the corresponding 2 beat delay lines TD L of the temperature interval number prediction model, the output of the 2 beat delay lines TD L are respectively used as the input of the corresponding 2 upper limit value and lower limit value ANFIS neural network prediction models, the difference between the upper limit value and the lower limit value of the output interval number of the highway multipoint temperature fusion model and the output of the corresponding 2 upper limit value and lower limit value ANFIS neural network prediction models is respectively the input of the corresponding 2 upper limit value L STM neural network residual prediction models, the output of the 2 upper limit value ANFIS neural network prediction models and the output of the 2 upper limit value L STM neural network residual prediction models are used as the input of the interval number Elman neural network fusion model, and the output of the Elman neural network fusion model is the predicted value of the temperature interval number detected in a period of a;
① ANFIS neural network prediction model design
The ANFIS neural network prediction model is an Adaptive Fuzzy Inference System ANFIS based on a neural network, also called an Adaptive neural Fuzzy Inference System (Adaptive neural-Fuzzy Inference System), organically combines the neural network and the Adaptive Fuzzy Inference System, can exert the advantages of the neural network and the Adaptive Fuzzy Inference System, and can make up the respective defects, a Fuzzy membership function and a Fuzzy rule in the ANFIS neural network prediction model are obtained by learning known historical data of a large number of road temperatures, the ANFIS neural network prediction model with the largest characteristic of interval number is a data-based modeling method instead of any given interval based on experience or intuition, the input of the ANFIS neural network prediction model is 2 outputs according to a time delay line TD L, the output of the 2 ANFIS neural network prediction models is a predicted value of upper limit and lower limit values of the temperature interval number, and the main operation steps are as follows, the layer 1, the upper limit value and the lower limit value of the temperature interval value in a period of time delay input is expressed as the corresponding output of each road node:
Figure BDA0002419625480000151
the formula n is the number of each input membership function, and the membership function adopts a Gaussian membership function.
And 2, realizing rule operation, outputting the applicability of the rule, and multiplying the rule operation of the ANFIS neural network by adopting multiplication.
Figure BDA0002419625480000152
And 3, normalizing the applicability of each rule:
Figure BDA0002419625480000153
and 4, at the layer 4, the transfer function of each node is a linear function and represents a local linear model, and the output of each self-adaptive node i is as follows:
Figure BDA0002419625480000154
and 5, a single node of the layer is a fixed node, and the output of the ANFIS neural network prediction model is calculated as follows:
Figure BDA0002419625480000155
the condition parameters determining the shape of the membership function and the conclusion parameters of the inference rule in the ANFIS neural network prediction model can be trained through a learning process. The parameters are adjusted by an algorithm combining a linear least square estimation algorithm and gradient descent. In each iteration of the ANFIS neural network prediction model, firstly, input signals are transmitted to the layer 4 along the network in the forward direction, and conclusion parameters are adjusted by adopting a least square estimation algorithm; the signal continues to propagate forward along the network to the output layer (i.e., layer 5). The ANFIS neural network prediction model reversely propagates the obtained error signals along the network, and the condition parameters are updated by a gradient method. By adjusting the given condition parameters in the ANFIS neural network in this way, the global optimum point of the conclusion parameters can be obtained, so that the dimension of the search space in the gradient method can be reduced, and the convergence speed of the ANFIS neural network parameters can be increased. The output of the 2 ANFIS neural network prediction models is the predicted value of the upper limit value and the lower limit value of the temperature interval number.
② and L STM neural network residual prediction model design
RoadThe difference between the upper and lower limit values of the multi-point temperature fusion model Output interval number and the corresponding 2 upper and lower limit values ANFIS neural network prediction model Output is respectively the Input of the corresponding 2 upper and lower limit values L STM neural network residual prediction model, the 2 upper and lower limit values L STM neural network residual prediction model Output is the Input of the interval number Elman neural network fusion model, the L STM neural network residual prediction model is a time Recursive Neural Network (RNN) consisting of long and short term Memory (L1 STM) units, the time Recursive Neural Network (RNN) is also generally referred to as L STM time recursive neural network, the 462 STM neural network residual prediction model introduces mechanisms of Memory cells (Memory cells) and hidden layer states (Cell states) to control information transfer between hidden layers, the Memory cells of an L STM neural network have 3 Gate(s) calculation structures respectively of Input Gates (Input Gates), forgetting Gates (Gates) and Output Gates (Gate) and the Output Gates of the hidden Gate (STM) are respectively Input Gates, the forgetting Gate (Gate) or the Input Gate) can be used for controlling the addition of new or the addition of the Input Gate (Input Gate) of the Input of the linear neural network residual information and the hidden Gate (Gate) of the hidden Gate (Gate) when the hidden Gate structure is considered to be a long-based on the long-term neural network, the hidden-forgetting-learning-based on-learning-missing-short-missing-short-time Memory unit, the long-short1,x2,…,xT) The hidden layer state is (h)1,h2,…,hT) Then, time t has:
it＝sigmoid(Whiht-1+WxiXt) (26)
ft＝sigmoid(Whfht-1+WhfXt) (27)
ct＝ft⊙ct-1+it⊙tanh(Whcht-1+WxcXt) (28)
ot＝sigmoid(Whoht-1+WhxXt+Wcoct) (29)
ht＝ot⊙tanh(ct) (30)
wherein it、ft、otRepresenting input, forget and output doors, ctRepresenting cell cells, Wh represents the weight of the recursive connections,
the method comprises the steps of firstly establishing L STM neural network residual prediction models, establishing training sets by using residual data of predicted values of the upper limit value and the lower limit value of the preprocessed temperature interval number, training the models, and considering the time sequence and nonlinearity of the residual data of the predicted values of the upper limit value and the lower limit value of the temperature interval number by L STM neural network residual prediction models, so that the prediction precision is high.
③ and interval number Elman neural network fusion model design
The output of 2 upper and lower limit value ANFIS neural network prediction models and the output of 2 upper and lower limit value L STM neural network residual prediction models are used as the input of an interval number Elman neural network fusion model, the output of the interval number Elman neural network fusion model is the predicted value of the detected temperature interval number in a period of timeA special association layer; the correlation layer receives the feedback signal from the hidden layer, and each hidden layer node is connected with the corresponding correlation layer node. The association layer takes the hidden layer state at the previous moment and the network input at the current moment as the input of the hidden layer, which is equivalent to state feedback. The transfer function of the hidden layer is generally a Sigmoid function, the output layer is a linear function, and the associated layer is also a linear function. In order to effectively solve the problem of approximation accuracy in road temperature fusion prediction, the function of a correlation layer is enhanced. Setting the numbers of an input layer, an output layer and a hidden layer of the Elman neural network fusion model as m, n and r respectively; w is a1，w2，w3And w4Respectively representing the connection weight matrixes from the structural layer unit to the hidden layer, from the input layer to the hidden layer, from the hidden layer to the output layer and from the structural layer to the output layer, and then the expressions of the hidden layer, the associated layer and the output layer of the Elman neural network fusion model are respectively as follows:
Figure BDA0002419625480000181
cp(k)＝xp(k-1) (32)
Figure BDA0002419625480000182
the number of an input layer, an output layer and a hidden layer of each Elman neural network fusion model is respectively 4,2 and 11, the input of the model is the predicted values of the upper limit and the lower limit of the temperature interval number and the predicted value of the residual error of the upper limit and the lower limit, and the input of the model is the predicted value of the temperature interval number, so that the prediction of the temperature interval number is realized, and the prediction accuracy is improved.
(4) Design of interval number least square support vector machine L S-SVM road surface slippery wet classifier
The method comprises the following steps that a least square support vector machine L S-SVM with the input of a region number least square support vector machine L S-SVM road surface slippery-wet classifier being 3 region numbers and the output of a region number 1 is a least square support vector machine L S-SVM, the output of a temperature detection module, a rain and snow detection module and an icing detection module is the input of a region number least square support vector machine L S-SVM road surface slippery-wet classifier, the output of a region number least square support vector machine L S-SVM road surface slippery-wet classifier is the region number representing the magnitude of the slippery degree grade of a detected public road, according to the influence engineering practice of slippery road surface driving caused by slippery wet on the road surface and a road surface grade and surface layer type (GB/T-920), the region number support vector machine L S-SVM road surface slippery-wet classifier, a relation table 1 of 5 region numbers corresponding to 5 road surface slippery degree grades is constructed, the 5 road surface slippery degree grade of slippery degree is divided into a dry state, slippery, the slippery condition with slippery, the slippery condition, the danger is compared, the danger is very poor and the danger is solved, the problem that the most approximate to the optimal linear regression of the smooth slippery condition:
Figure BDA0002419625480000191
in the solving process, in order to avoid solving a complex nonlinear mapping function, a Radial Basis Function (RBF) is introduced to replace dot product operation in a high-dimensional space, so that the calculated amount can be greatly reduced, and the RBF kernel function is easy to realize the optimization process of the SVM, because the center of each basis function of the RBF kernel function corresponds to the support vector one by one, and the support vector and the weight can be obtained through an algorithm, therefore, the interval number least square support vector machine L S-SVM road surface slip-moisture classifier is as follows:
Figure BDA0002419625480000192
j is 2 to represent the interval number formed by 2 nodes, and the interval number is the interval number of which the output of the interval number least square support vector machine L S-SVM road surface slippery wet classifier represents the magnitude of the slippery wet degree grade of the detected road.
TABLE 1 highway pavement slippery degree grade and interval number corresponding relation table
Serial number Degree of danger of road hydroplaning Number of intervals
1 Dry state [0.00，0.20]
2 Slippery and wet tracing [0.20，0.40]
3 Is relatively dangerous [0.40，0.60]
4 Is very dangerous [0.60，0.80]
5 Is very dangerous [0.80，1.0]
5. Design example of highway meteorological environment parameter acquisition and measurement platform
According to the condition of the highway meteorological environment, the system designs a plane layout installation diagram of a detection node 1, a control node 2 and a field monitoring terminal 3, wherein the detection node 1 is arranged in the detected highway meteorological environment in a balanced mode, and the system is used for achieving the collection of the highway meteorological environment parameters and the detection and prediction of the highway meteorological environment parameters.
The technical means disclosed in the invention scheme are not limited to the technical means disclosed in the above embodiments, but also include the technical scheme formed by any combination of the above technical features. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and such improvements and modifications are also considered to be within the scope of the present invention.

Claims (8)

1. An intelligent detection system for road meteorological parameters and road surface slippery conditions is characterized in that the detection system consists of a road meteorological environment parameter acquisition platform based on a CAN bus and a road surface slippery condition grade classification system, the road meteorological environment parameter acquisition platform based on the CAN bus realizes the detection and the monitoring of road meteorological environment factor parameters, the road surface slippery condition grade classification system consists of a temperature detection module, a rain and snow detection module, an icing detection module and an interval number least square support vector machine L S-SVM road surface slippery condition classifier, the outputs of the temperature detection module, the rain and snow detection module and the icing detection module are used as the inputs of an interval number least square support vector machine L S-SVM road surface slippery condition classifier, the output of the interval number least square support vector machine L S-SVM road surface slippery condition grade classifier is an interval number representing the detected road surface slippery condition grade, and the road surface slippery condition grade classification system realizes the detection and the classification of the road surface slippery condition grade;
the temperature detection module consists of a plurality of temperature interval number neural network models, a highway multipoint temperature fusion model and a temperature interval number prediction model, the output of all detection point temperature sensors is used as the input of the corresponding temperature interval number neural network models, the output of the temperature interval number neural network models is used as the input of the highway multipoint temperature fusion model, the output of the highway multipoint temperature fusion model is used as the input of the temperature interval number prediction model, and the output of the temperature interval number prediction model is used as the input of an interval number least square support vector machine L S-SVM road surface slippery moisture classifier;
the section number least square support vector machine L S-SVM road surface slippery wet classifier is a least square support vector machine L S-SVM with the input of 3 sections and the output of 1 section, the output of the temperature detection module, the rain and snow detection module and the icing detection module is the input of a section number least square support vector machine L S-SVM road surface slippery wet classifier, the output of the section number least square support vector machine L S-SVM road surface slippery wet classifier is the section number representing the magnitude of the slippery wet degree grade of a detected road, according to engineering practice of slippery road surface on driving and national standards for road surface slippery wet determination, the section number support vector machine L S-SVM road surface slippery wet classifier constructs a corresponding relation table of 5 slippery wet degree grades and 5 section numbers of the road surface, the 5 slippery wet degree grades of the road surface are respectively in a dry state, have slippery wet signs, are relatively dangerous, are very dangerous and very dangerous, the calculation section number least square support vector machine L S-SVM road surface slippery wet classifier outputs the most square support vector machine L S-SVM road surface slippery wet classifier, and the output of the section number least square support vector machine L S-SVM road surface slippery wet classifier are determined as the input of the section number.
2. The system as claimed in claim 1, wherein the neural network model for temperature interval number is composed of RR time recurrent neural network, DRNN neural network model for interval number and 3 beat delay lines TD L, the neural network model for temperature interval number converts the temperature value of the measured road sensed by the road temperature sensor in a period of time into dynamic interval value of road temperature, the output of each detected point temperature sensor is the input of the corresponding RR time recurrent neural network, the output of the RR time recurrent neural network is the input of the corresponding 1 beat delay line TD L, the output of the 3 beat delay lines TD L is the input of the DRNN neural network model for interval number, the output of the DRNN neural network model is the output of the neural network model for temperature interval number and the interval number formed by the upper and lower limit values representing the road temperature in a period of time, and the upper and lower limit values of the interval number of road temperature number are respectively used as the input of the corresponding 2 beat delay lines TD L.
3. The intelligent road meteorological parameter and road surface slippery detection system as claimed in claim 1, wherein: the road multipoint temperature fusion model forms a time sequence temperature interval number array according to the output of the temperature interval number neural network model of all the detection points in a period of time, determines the positive and negative ideal values of the time sequence temperature interval number array, and respectively calculates the distance and similarity between the time sequence temperature interval value of each detection point and the positive and negative ideal values of the time sequence temperature interval number array; and the quotient obtained by dividing the negative ideal value distance of the time-series temperature interval numerical value of each detection point by the sum of the negative ideal value distance of the time-series temperature interval numerical value of each detection point and the positive ideal value distance of the time-series temperature interval numerical value of each detection point is the distance relative closeness of the time-series temperature interval numerical value of each detection point, and the quotient obtained by dividing the distance relative closeness of the time-series temperature interval numerical value of each detection point by the sum of the distance relative closeness of the time-series temperature interval numerical values of all the detection points is the distance fusion weight of the time-series temperature interval numerical value of each detection point.
4. The system for intelligently detecting the meteorological parameters of the highway and the slippery road surface as claimed in claim 1, wherein the temperature interval number prediction model consists of 2 beat-to-beat delay lines TD L, 2 ANFIS neural network prediction models, 2L STM neural network residual error prediction models and an interval number Elman neural network fusion model, the upper limit value and the lower limit value of the output interval number of the highway multipoint temperature fusion model are respectively the corresponding 2 beat-to-beat delay line TD L inputs of the temperature interval number prediction model, the 2 beat-to-beat delay line TD L outputs are respectively the corresponding 2 upper limit value and lower limit value ANFIS neural network prediction models, the difference between the upper limit value and the lower limit value of the output interval number of the highway multipoint temperature fusion model and the corresponding 2 upper limit value and lower limit value ANFIS neural network prediction model outputs is respectively the corresponding 2 upper limit value and lower limit value L neural network residual error prediction models, the 2 upper limit value ANFIS neural network prediction model outputs and the 2 upper limit value L STM neural network residual error prediction models are respectively the corresponding 2 upper limit value of the input interval number Elman neural network residual error prediction model, and the detected temperature interval number Elman neural network prediction model outputs are a section of the temperature prediction model, and the detected temperature interval number Elman neural network.
5. The intelligent road meteorological parameter and road surface slippery detection system of claim 3, wherein: the quotient obtained by dividing the positive ideal value similarity of the time-series temperature interval numerical value of each detection point in a period of time by the sum of the negative ideal value similarity of the time-series temperature interval numerical value of each detection point and the positive ideal value similarity of the time-series temperature interval numerical value of each detection point is the relative closeness of the time-series temperature interval numerical values of each detection point, and the quotient obtained by dividing the relative closeness of the similarity of the time-series temperature interval numerical values of each detection point by the sum of the relative closeness of the similarity of the time-series temperature interval numerical values of all the detection points is the similarity fusion weight of the time-series temperature interval numerical values of each detection point;
the ratio of the distance fusion weight of the time sequence temperature interval numerical value of each detection point of the highway in a period of time to the root mean square of the similarity fusion weight product to the distance fusion weight of the time sequence temperature interval numerical values of all the temperature detection points and the root mean square sum of the similarity fusion weight product is the root mean square combination weight of the time sequence temperature interval numerical value fusion of the detection points; the distance fusion weight and the similarity fusion weight of the time sequence temperature interval numerical value of each detection point in a period of time are linearly combined into the linear combination weight of the time sequence temperature interval numerical value of the highway temperature detection point, the root mean square combination weight and the linear combination weight of the time sequence temperature interval numerical value fusion of each highway temperature detection point form the time sequence temperature interval numerical value fusion weight of the detection point, and the sum of the time sequence temperature interval numerical value of each detection point sensor and the time sequence temperature interval fusion weight of the detection point sensor is the fusion value of the time sequence interval numerical values of all the temperature detection points of the highway.
6. The intelligent road meteorological parameter and road surface slippery detection system as claimed in claim 1, wherein: the CAN bus-based highway meteorological environment parameter acquisition platform consists of a detection node, a control node and an on-site monitoring terminal, and communication among the detection node, the control node and the on-site monitoring terminal is realized through the CAN bus; the detection nodes are respectively composed of a sensor group module, a single chip microcomputer and a communication interface, the sensor group module is responsible for detecting the temperature, rain and snow, icing and visibility highway meteorological environment parameters of the highway meteorological environment, the sampling interval is controlled by the single chip microcomputer, and the sampling interval is sent to the field monitoring end through the communication module.
7. The intelligent road meteorological parameter and road surface slippery detection system of claim 6, wherein: the control node controls the adjusting equipment of the highway meteorological environment parameters; the field monitoring end consists of an industrial control computer and an RS232/CAN communication module, and is used for managing meteorological environment parameters of the detection nodes for detecting the highway and constructing a road surface slippery degree grade classification system.
8. The intelligent road meteorological parameter and road surface slippery detection system as claimed in claim 1, wherein: the rain and snow detection module consists of a plurality of rain and snow interval number neural network models, a multi-point rain and snow fusion model and a rain and snow interval number prediction model; the icing detection module consists of a plurality of icing interval number neural network models, a road multi-point icing fusion model and an icing interval number prediction model.
CN202010201762.3A 2020-03-20 2020-03-20 Intelligent detection system for highway meteorological parameters and road surface slippery Withdrawn CN111476463A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010201762.3A CN111476463A (en) 2020-03-20 2020-03-20 Intelligent detection system for highway meteorological parameters and road surface slippery

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010201762.3A CN111476463A (en) 2020-03-20 2020-03-20 Intelligent detection system for highway meteorological parameters and road surface slippery

Publications (1)

Publication Number Publication Date
CN111476463A true CN111476463A (en) 2020-07-31

Family

ID=71747715

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010201762.3A Withdrawn CN111476463A (en) 2020-03-20 2020-03-20 Intelligent detection system for highway meteorological parameters and road surface slippery

Country Status (1)

Country Link
CN (1) CN111476463A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112904756A (en) * 2021-01-13 2021-06-04 淮阴工学院 Pipe network big data detection system
CN114970745A (en) * 2022-06-17 2022-08-30 淮阴工学院 Intelligent security and environment big data system of Internet of things

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112904756A (en) * 2021-01-13 2021-06-04 淮阴工学院 Pipe network big data detection system
CN114970745A (en) * 2022-06-17 2022-08-30 淮阴工学院 Intelligent security and environment big data system of Internet of things

Similar Documents

Publication Publication Date Title
CN111461413B (en) Highway road surface performance detecting system
CN111429718B (en) Intelligent detection system for road traffic safety
CN110580021B (en) Granary environmental safety intelligent monitoring system based on field bus
CN109859469A (en) A kind of vehicle flowrate prediction technique based on integrated LSTM neural network
CN111460729A (en) Bridge deformation intelligent detection system
CN110705757B (en) Multi-point temperature sensor intelligent monitoring system based on field bus network
CN111444947B (en) Bridge settlement intelligent detection system
CN110766132B (en) Orchard yield intelligent prediction system based on Internet of things
CN112665656B (en) Big data detection system of agricultural product growth environment
CN107516145A (en) A kind of multichannel photovoltaic power generation output forecasting method based on weighted euclidean distance pattern classification
CN111461187B (en) Intelligent building settlement detection system
CN114397043B (en) Multi-point temperature intelligent detection system
CN111476463A (en) Intelligent detection system for highway meteorological parameters and road surface slippery
CN112881601B (en) Moisture detecting system based on cloud platform
CN107862877A (en) A kind of urban traffic signal fuzzy control method
CN116167527B (en) Pure data-driven power system static safety operation risk online assessment method
CN107436409A (en) A kind of electric automobile power battery SOC intelligent predicting devices
CN111474094B (en) Intelligent dust concentration detection system
CN113536665B (en) Road surface temperature short-term prediction method and system based on characteristic engineering and LSTM
CN115062764B (en) Intelligent illuminance adjustment and environmental parameter Internet of things big data system
CN114386672B (en) Environment big data Internet of things intelligent detection system
Peng et al. Meteorological satellite operation prediction using a BiLSTM deep learning model
CN114970745B (en) Intelligent security and environment big data system of Internet of things
CN110827134A (en) Power grid enterprise financial health diagnosis method
CN114399024B (en) Oil gas concentration big data intelligent detection system

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
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20200731