Energy-saving control system and method for road operation risk prevention and control
1. The utility model provides an energy-saving control system of road operation risk prevention and control which characterized in that includes: the system comprises traffic monitoring equipment, a cloud computing platform and a variable information board which are arranged on the road side; the traffic flow monitoring equipment comprises a radar, a traffic detector and a visual sensor;
the radar is used for detecting the speed of a vehicle on a road and transmitting the speed to the cloud computing platform;
the traffic detector is used for detecting traffic events on a road in real time and transmitting the traffic events to the cloud computing platform;
the vision sensor is used for acquiring the running track of the vehicle on the road in real time and transmitting the running track to the cloud computing platform;
the cloud computing platform determines the traffic abnormal grade and issues a corresponding prevention and control strategy according to the vehicle speed, the traffic event and the running track of the vehicle;
the variable information board is arranged at the road side of the road and used for displaying the speed limit value on the road section according to the control strategy.
2. An energy-saving control method for preventing and controlling road running risks is characterized by comprising the following steps:
step 1, equally dividing a highway into a plurality of sub-road sections, wherein each sub-road section is provided with a speed limiting node, each speed limiting node is provided with a speed limiting device, each speed limiting device corresponds to a speed limiting state, and each sub-road section is provided with a traffic monitoring device to obtain the speed, traffic events and vehicle running tracks of vehicles on the sub-road sections;
the method comprises the following steps of respectively acquiring the speed, traffic events and running tracks of a vehicle through a radar, a traffic detector and a visual sensor which are installed on the roadside;
step 2, predicting the road section risks on each sub road section according to the speed, traffic events and vehicle running tracks of the vehicles on all the sub road sections;
step 3, acquiring traffic flow on the sub-road sections, loading road section risks on each sub-road section onto the traffic flow, and predicting the space-time level distribution of the road section risks on each sub-road section according to the conduction evolution rule of the traffic flow on the road section;
step 4, determining the speed limit value of the speed limit equipment on each sub-road section according to the space-time grade distribution of the road section risks on each sub-road section;
and 5, determining the working state of the downstream variable speed-limiting plate according to the speed-limiting value of the speed-limiting equipment on each sub-road section.
3. The energy-saving control method for road operation risk prevention and control according to claim 2, wherein the state of the speed limiting device is specifically as follows:
if a static speed limiting plate is arranged on the sub-road section, the state of the speed limiting equipment corresponding to the speed limiting node on the road section is 1;
if the variable speed-limiting plate is arranged on the sub-road section, the state of the speed-limiting equipment corresponding to the speed-limiting node on the road section is 2.
4. The energy-saving control method for road operation risk prevention and control according to claim 2, wherein the predicting of the road segment risk on each sub-road segment is specifically:
(2.1) establishing a Bayes deep learning model, and training and testing the Bayes deep learning model by using a Bayes statistical method;
(2.2) acquiring a training sample set and a testing sample set, and training a Bayes deep learning model through the training sample to obtain a trained Bayes deep learning model; testing the trained Bayes deep learning model through the test sample, and determining that the Bayes deep learning model is trained when the average absolute error between the output of the Bayes deep learning model and the label of the test sample is less than a preset error threshold;
the training sample set and the testing sample set are respectively selected from historical samples, and each sample is the speed of a vehicle, a traffic event and a vehicle running track;
and (2.3) taking the currently acquired vehicle speed, traffic events and vehicle running tracks of the vehicles on each sub-road section as input data of the trained Bayes deep learning model, and outputting road section risks on each sub-road section through the training and learning of the trained Bayes deep learning model.
5. The energy-saving control method for road operation risk prevention and control according to claim 2, wherein the temporal-spatial level distribution of the road segment risk on each sub-road segment is predicted, specifically: the risk of the road section is divided into five grades, namely five grades of low risk, general risk, medium risk, larger risk and important risk, wherein the risk occurrence probability intervals corresponding to the low risk, the general risk, the medium risk, the larger risk and the important risk are respectively (0, 5% ], (5%, 20% ], (20%, 50% ], (50%, 80% ] and (80%, 100% ]).
6. The energy-saving control method for road operation risk prevention and control according to claim 5, wherein the risk occurrence probability is calculated as follows:
wherein (S)1,S2,S3,……Sn) To predict a set of factors of traffic risk A, where S1,S2,S3,……SnAre independent of each otherAnd (4) variable quantity.
7. The energy-saving control method for road operation risk prevention and control according to claim 2, wherein the speed limit value of the speed limit device on each sub-section is determined, specifically:
when the risk occurrence probability interval is (0, 5% >), taking the speed limit value as 120 km/h;
when the risk occurrence probability interval (5%, 20% >) is determined, the speed limit value is taken as 100 km/h;
when the risk occurrence probability interval (20%, 50% >) is determined, the speed limit value is 80 km/h;
when the risk occurrence probability interval (50%, 80%) is reached, the speed limit value is 60 km/h;
when the risk occurrence probability interval (80%, 100%) is reached, the speed limit value is less than 60 km/h;
8. the energy-saving control method for road operation risk prevention and control according to claim 2, wherein the working state of the downstream variable speed-limiting plate is determined, specifically:
when the speed limit value of the speed limit equipment on each sub-road section is equal to the speed limit value of the speed limit equipment on the upstream sub-road section, the variable speed limit plate on the sub-road section enters a sleep mode; otherwise, the variable speed limit board on the sub-road section enters a working mode.
Background
In recent years, highway construction mileage and automobile holding capacity have increased dramatically, and the number of vehicles per year has changed, so that the traffic flow has also changed. The traditional static speed limit sign has strong constraint and cannot adapt to the change of traffic flow. The variable speed limit control technology becomes a traffic control technology which is widely concerned by various countries. The variable speed-limiting system can make real-time reaction to the changing traffic environment characteristics and automatically adjust the current speed-limiting value. Therefore, traffic flow is guided, the traffic network is ensured to be safe and smooth, and traffic accidents are reduced and the loss caused by the traffic accidents is reduced.
At present, a great number of variable speed-limiting boards are not used on the highway, one reason is high cost, difficult power supply and high power consumption, and the existing variable speed-limiting boards are mainly used for risk prevention and control and are not provided with energy-saving variable information boards.
Disclosure of Invention
Aiming at the problems in the prior art, the invention aims to provide an energy-saving control system and method for road operation risk prevention and control, which can selectively and temporarily close some variable speed-limiting plates under the condition of ensuring risk prevention and control so as to realize energy-saving control on the variable speed-limiting plates.
In order to achieve the purpose, the invention adopts the following technical scheme:
the utility model provides an energy-conserving control system of road operation risk prevention and control, includes: the system comprises traffic monitoring equipment, a cloud computing platform and a variable information board which are arranged on the road side; the traffic flow monitoring equipment comprises a radar, a traffic detector and a visual sensor;
the radar is used for detecting the speed of a vehicle on a road and transmitting the speed to the cloud computing platform;
the traffic detector is used for detecting traffic events on a road in real time and transmitting the traffic events to the cloud computing platform;
the vision sensor is used for acquiring the running track of the vehicle on the road in real time and transmitting the running track to the cloud computing platform;
the cloud computing platform determines the traffic abnormal grade and issues a corresponding prevention and control strategy according to the vehicle speed, the traffic event and the running track of the vehicle;
the variable information board is arranged at the road side of the road and used for displaying the speed limit value of the road section according to the control strategy.
(II) an energy-saving control method for preventing and controlling road operation risks comprises the following steps:
step 1, equally dividing a highway into a plurality of sub-road sections, wherein each sub-road section is provided with a speed limiting node, each speed limiting node is provided with a speed limiting device, each speed limiting device corresponds to a speed limiting state, and each sub-road section is provided with a traffic monitoring device to obtain the speed, traffic events and vehicle running tracks of vehicles on the sub-road sections;
the method comprises the following steps of respectively acquiring the speed, traffic events and running tracks of a vehicle through a radar, a traffic detector and a visual sensor which are installed on the roadside;
step 2, predicting the road section risks on each sub road section according to the speed, traffic events and vehicle running tracks of the vehicles on all the sub road sections;
step 3, acquiring traffic flow on the sub-road sections, loading road section risks on each sub-road section onto the traffic flow, and predicting the space-time level distribution of the road section risks on each sub-road section according to the conduction evolution rule of the traffic flow on the road section;
step 4, determining the speed limit value of the speed limit equipment on each sub-road section according to the space-time grade distribution of the road section risks on each sub-road section;
and 5, determining the working state of the downstream variable speed-limiting plate according to the speed-limiting value of the speed-limiting equipment on each sub-road section.
Further, the state of the speed limiting device is specifically as follows:
if a fixed speed limit sign is arranged on the sub-road section, the state of the speed limit equipment corresponding to the speed limit node on the road section is 1;
if the variable speed limit sign is set on the sub-road section, the state of the speed limit device corresponding to the speed limit node on the road section is 2.
Further, predicting the road segment risk on each sub-road segment specifically includes:
(2.1) establishing a Bayes deep learning model, and training and testing the Bayes deep learning model by using a Bayes statistical method;
(2.2) acquiring a training sample set and a testing sample set, and training a Bayes deep learning model through the training sample to obtain a trained Bayes deep learning model; testing the trained Bayes deep learning model through the test sample, and determining that the Bayes deep learning model is trained when the average absolute error between the output of the Bayes deep learning model and the label of the test sample is smaller than a preset error threshold, wherein the preset error threshold is set according to the precision of the Bayes deep learning model in the actual prediction process, and the method is not specifically limited herein;
the training sample set and the testing sample set are respectively selected from historical samples, and each sample is the speed of a vehicle, a traffic event and a vehicle running track;
and (2.3) taking the currently acquired vehicle speed, traffic events and vehicle running tracks of the vehicles on each sub-road section as input data of the trained Bayes deep learning model, and outputting road section risks on each sub-road section through the training and learning of the trained Bayes deep learning model.
Further, predicting the space-time level distribution of the road section risks on each sub-road section specifically comprises: the risk of the road section is divided into five grades, namely five grades of low risk, general risk, medium risk, larger risk and important risk, wherein the risk occurrence probability intervals corresponding to the low risk, the general risk, the medium risk, the larger risk and the important risk are respectively (0, 5% ], (5%, 20% ], (20%, 50% ], (50%, 80% ] and (80%, 100% ]).
Further, the risk occurrence probability is calculated as follows:
wherein (S)1,S2,S3,……Sn) To predict a set of factors of traffic risk A, where S1,S2,S3,……SnAre variables that are independent of each other.
Further, determining the speed limit value of the speed limit equipment on each sub-road section specifically comprises:
when the risk occurrence probability interval is (0, 5% >), taking the speed limit value as 120 km/h;
when the risk occurrence probability interval (5%, 20% >) is determined, the speed limit value is taken as 100 km/h;
when the risk occurrence probability interval (20%, 50% >) is determined, the speed limit value is 80 km/h;
when the risk occurrence probability interval (50%, 80%) is reached, the speed limit value is 60 km/h;
when the risk occurrence probability interval (80%, 100%) is reached, the speed limit value is less than 60 km/h;
further, determining the working state of the downstream variable speed-limiting plate specifically comprises the following steps:
when the speed limit value of the speed limit equipment on each sub-road section is equal to the speed limit value of the speed limit equipment on the upstream sub-road section, the variable speed limit plate on the sub-road section enters a sleep mode; otherwise, the variable speed limit board on the sub-road section enters a working mode.
Compared with the prior art, the invention has the beneficial effects that:
on the basis of risk prevention and control, the risk grade on each sub-road section is predicted according to a risk conduction evolution rule, the space-time grade distribution of the road section risk on each sub-road section is predicted according to the risk grade on each sub-road section, the speed limit value corresponding to the speed limit equipment on each sub-road section is determined according to the space-time grade distribution of the road section risk on each sub-road section, then the speed limit value of the speed limit equipment on each sub-road section is compared with the speed limit value on the upstream sub-road section, and when the speed limit value on the variable speed limit board on each sub-road section is equal to the speed limit value on the upstream sub-road section, the variable speed limit board on the sub-road section is closed, so that the energy-saving effect is achieved.
Drawings
Fig. 1 is a structural diagram of an energy-saving control system for road running risk prevention and control according to embodiment 1 of the present invention;
fig. 2 is a flowchart of an energy-saving control method for road operation risk prevention and control according to embodiment 2 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. Test methods in which specific conditions are not specified in the following examples are generally carried out under conventional conditions or under conditions recommended by the respective manufacturers.
Example 1
Referring to fig. 1, an energy-saving control system for road operation risk prevention and control includes: the system comprises traffic monitoring equipment, a cloud computing platform and a variable information board which are arranged on the road side; the traffic flow monitoring equipment comprises a radar, a traffic detector and a visual sensor;
the radar is used for detecting the speed of a vehicle on a road and transmitting the speed to the cloud computing platform;
the traffic detector is used for detecting traffic events on a road in real time and transmitting the traffic events to the cloud computing platform;
the vision sensor is used for acquiring the running track of the vehicle on the road in real time and transmitting the running track to the cloud computing platform;
the cloud computing platform determines the traffic abnormal grade and issues a corresponding prevention and control strategy according to the vehicle speed, the traffic event and the running track of the vehicle;
the variable information board is arranged at the road side of the road and used for displaying the speed limit value of the road section according to the control strategy.
Example 2
Referring to fig. 2, an energy-saving control method for road operation risk prevention and control includes the following steps:
step 1, equally dividing a highway into a plurality of sub-road sections, wherein each sub-road section is provided with a speed limiting node, each speed limiting node is provided with a speed limiting device, each speed limiting device corresponds to a speed limiting state, and each sub-road section is provided with a traffic monitoring device to obtain the speed, traffic events and vehicle running tracks of vehicles on the sub-road sections;
numbering each speed limit node in sequence along the running direction of the vehicle, wherein the specific number is a1，a2，a3，……，anN is the total number of speed limit nodes in the control area;
the method comprises the following steps of respectively acquiring the speed, traffic events and running tracks of a vehicle through a radar, a traffic detector and a visual sensor which are installed on the roadside;
specifically, the states of the speed limiting device include the following two states:
1) if a static speed limiting plate is arranged on the sub-road section, the state of the speed limiting equipment corresponding to the speed limiting node on the road section is 1;
2) if the variable speed-limiting plate is arranged on the sub-road section, the state of the speed-limiting equipment corresponding to the speed-limiting node on the road section is 2;
that is, the speed limiting device on each sub-road segment is a dynamic speed limiting board or a static speed limiting board.
Step 2, predicting the road section risks on each sub road section according to the speed, traffic events and vehicle running tracks of the vehicles on all the sub road sections;
specifically, predicting the road segment risk on each sub-road segment specifically includes:
(2.1) establishing a Bayes deep learning model, and training and testing the Bayes deep learning model by using a Bayes statistical method;
specifically, a Bayesian statistical method is used for carrying out parameter estimation on parameters to be optimized of the deep learning model, wherein the parameters to be optimized comprise weight parameters and bias parameters among layers. A Bayesian deep learning model for training parameter optimization using the training set;
further, the specific process of using the bayesian statistical method to perform parameter estimation on the parameter to be optimized of the deep learning model is as follows:
firstly, setting the prior probability of a parameter to be optimized to obey standard normal distribution, and determining the posterior probability distribution of the parameter to be optimized through the likelihood and the prior probability;
secondly, sampling the distribution of the weight parameters and the bias parameters for multiple times by using an MCMC method to obtain a parameter combination set;
and finally, training the Bayesian deep learning model pair by using the parameter combination set and the training data set.
The Bayesian depth learning model comprises a dropout layer, and a Relu activation function is used for fitting data in a training data set and a parameter combination set;
(2.2) acquiring a training sample set and a testing sample set, and training a Bayes deep learning model through the training sample to obtain a trained Bayes deep learning model; testing the trained Bayes deep learning model through the test sample, and determining that the Bayes deep learning model is trained when the average absolute error between the output of the Bayes deep learning model and the label of the test sample is less than a preset error threshold;
the training sample set and the testing sample set are respectively selected from historical samples, and each sample is the speed of a vehicle, a traffic event and a vehicle running track;
(2.3) taking the currently acquired vehicle speed, traffic events and vehicle running tracks of the vehicles on each sub-road section as input data of the trained Bayes deep learning model, and outputting road section risks on each sub-road section through the training and learning of the trained Bayes deep learning model;
step 3, acquiring traffic flow information on the sub-road sections, loading road section risks on each sub-road section onto the traffic flow, and predicting the space-time level distribution of the road section risks on each sub-road section according to the conduction evolution rule of the traffic flow on the road section;
specifically, predicting the spatiotemporal level distribution of the road segment risks on each sub-road segment specifically includes: dividing the risk of the road section into five grades, namely a low risk grade, a general risk grade, a medium risk grade, a large risk grade and a major risk grade, wherein risk occurrence probability intervals corresponding to the low risk grade, the general risk grade, the medium risk grade, the large risk grade and the major risk grade are respectively (0, 5% ], (5%, 20% ], (20%, 50% ], (50%, 80% ] and (80%, 100%);
the division of the road section risk levels in the invention is based on a local standard 'urban road traffic congestion evaluation index system' from Beijing, and road section and regional traffic congestion indexes are respectively determined in the standard, wherein the road section congestion is divided into 5 levels by taking the average travel speed as a division basis; the congestion degree of the region is divided into five levels of very smooth, light congestion, moderate congestion and severe congestion by calculating the daily traffic congestion index of the region;
further, when the probability of congestion is high, it is indicated that the area or the road section is at a high risk, and when the probability of congestion is low, the road section or the area is at a low risk;
based on the above, the large-scale movable traffic jam risk is divided into five levels, namely low risk, general risk, medium risk, larger risk and major risk, by taking the jam occurrence probability as a risk grading index, and the five levels respectively correspond to risk occurrence probability intervals (0, 5%, (5%, 20%, (20%, 50%, (50%, 80% >) and (80%, 100%), and the risk occurrence probability interval corresponding to the traffic jam risk grade is according to a large-scale movable traffic risk early warning model based on a dynamic Bayesian network;
further, the risk occurrence probability is calculated as follows:
wherein (S)1,S2,S3,……Sn) To predict a set of factors of traffic risk A, where S1,S2,S3,……SnAre variables independent of each other;
specifically, the specific calculation process of the risk occurrence probability is as follows:
according to the condition independence:
finishing to obtain:
step 4, determining the speed limit value of the speed limit equipment on each sub-road section according to the space-time grade distribution of the road section risks on each sub-road section;
further, determining the speed limit value of the speed limit equipment on each sub-road section specifically comprises:
when the risk occurrence probability interval is (0, 5% >), taking the speed limit value as 120 km/h;
when the risk occurrence probability interval (5%, 20% >) is determined, the speed limit value is taken as 100 km/h;
when the risk occurrence probability interval (20%, 50% >) is determined, the speed limit value is 80 km/h;
when the risk occurrence probability interval (50%, 80%) is reached, the speed limit value is 60 km/h;
when the risk occurrence probability interval (80%, 100%) is reached, the speed limit value is less than 60 km/h;
specifically, the value of the speed limit value corresponding to the probability interval of the risk occurrence is according to the GB 238262009 expressway LED variable speed limit sign, the higher the risk occurrence probability is, the higher the probability of the risk occurrence is, the more the possibility of the risk occurrence is, the smaller the corresponding speed limit value is, so that the vehicle on the road section can be induced to run according to the issued speed limit value through the issuance of the speed limit value, the occurrence of traffic accidents is avoided, and the occurrence probability of the traffic accidents is reduced;
step 5, determining the working state of the downstream variable speed-limiting plate according to the speed-limiting value of the speed-limiting equipment on each sub-road section;
further, determining the working state of the downstream variable speed-limiting plate specifically comprises the following steps:
when the speed limit value of the speed limit equipment on each sub-road section is equal to the speed limit value of the speed limit equipment on the upstream sub-road section, the variable speed limit plate on the sub-road section enters a sleep mode; otherwise, the variable speed-limiting plate on the sub-road section enters a working mode;
the speed limiting equipment on each sub-road section is in two states, and the speed limiting equipment on each sub-road section is a static speed limiting plate or a dynamic speed limiting plate; when the static speed-limiting plate is arranged on the sub-road section, the speed-limiting value of the speed-limiting equipment determined in the step 4 is equal to the speed-limiting value on the sub-road section at the upstream of the sub-road section, the closing of the static speed-limiting plate is not considered, and the speed-limiting value on the static speed-limiting plate is determined; and when the dynamic speed-limiting plate is arranged on the sub-road section, the speed-limiting value of the speed-limiting equipment determined in the step 4 is equal to the speed-limiting value on the sub-road section at the upstream of the sub-road section, the variable speed-limiting plate on the sub-road section is closed, the energy-saving effect is realized, and meanwhile, the release of the speed-limiting value on the variable speed-limiting plate can induce the safe driving of the vehicle, so that the occurrence probability of traffic accidents is avoided.
To sum up, the energy-saving control method for road operation risk prevention and control provided by the invention detects the vehicle speed, traffic events and vehicle operation tracks of vehicles on each sub-road section every t seconds on the basis of risk prevention and control, predicts the risk level on each sub-road section according to the risk conduction evolution law, predicts the space-time level distribution of road section risk on each sub-road section according to the risk level on each sub-road section, determines the speed limit value corresponding to the speed limit equipment on each sub-road section according to the space-time level distribution of road section risk on each sub-road section, compares the speed limit value of the speed limit equipment on each sub-road section with the speed limit value on the upstream sub-road section, closes the variable speed limit plate on each sub-road section for t seconds when the speed limit value on the variable speed limit plate on each sub-road section is equal to the speed limit value on the upstream sub-road section, and the variable speed limit plate is circulated according to the working mode, so as to achieve the effect of energy saving.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
