CN117475628A - Expressway operation method and information system based on risk theory - Google Patents
Expressway operation method and information system based on risk theory Download PDFInfo
- Publication number
- CN117475628A CN117475628A CN202311441433.6A CN202311441433A CN117475628A CN 117475628 A CN117475628 A CN 117475628A CN 202311441433 A CN202311441433 A CN 202311441433A CN 117475628 A CN117475628 A CN 117475628A
- Authority
- CN
- China
- Prior art keywords
- road
- highway
- current
- road network
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000005457 optimization Methods 0.000 claims abstract description 311
- 238000013486 operation strategy Methods 0.000 claims abstract description 145
- 230000002159 abnormal effect Effects 0.000 claims abstract description 36
- 230000000007 visual effect Effects 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims abstract description 7
- 239000013598 vector Substances 0.000 claims description 175
- 230000006978 adaptation Effects 0.000 claims description 28
- 238000012544 monitoring process Methods 0.000 claims description 28
- 230000006835 compression Effects 0.000 claims description 21
- 238000007906 compression Methods 0.000 claims description 21
- 238000013210 evaluation model Methods 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 18
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000007418 data mining Methods 0.000 claims description 10
- 230000004927 fusion Effects 0.000 claims description 10
- 230000005484 gravity Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 8
- 230000003862 health status Effects 0.000 claims description 6
- 238000012423 maintenance Methods 0.000 claims description 6
- 230000008439 repair process Effects 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 230000000903 blocking effect Effects 0.000 claims description 2
- 238000012795 verification Methods 0.000 claims description 2
- 238000013461 design Methods 0.000 abstract description 5
- 230000036541 health Effects 0.000 abstract description 3
- 230000003044 adaptive effect Effects 0.000 abstract description 2
- 230000000875 corresponding effect Effects 0.000 description 45
- 230000008859 change Effects 0.000 description 13
- 206010039203 Road traffic accident Diseases 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 230000006399 behavior Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000009938 salting Methods 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000004148 unit process Methods 0.000 description 1
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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0133—Traffic data processing for classifying traffic situation
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
- G08G1/0141—Measuring and analyzing of parameters relative to traffic conditions for specific applications for traffic information dissemination
Landscapes
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a highway operation method and an information system based on a risk theory, wherein the method comprises the following steps: and acquiring the traffic flow information in the current road network through a road network real-time data acquisition interface, and identifying the running health state of the road. If abnormal road conditions exist, road condition optimization is carried out according to the current road optimization scheme. The optimized result and the corresponding operation strategy are obtained from a preset operation strategy library, and the pairing adaptability of the optimized result and the current highway network is analyzed. And when the pairing adaptability meets preset conditions, executing the operation strategy, and displaying an execution process on the visual large screen. By the design, the running condition of the highway network can be obtained accurately in real time, abnormal road conditions can be identified and processed rapidly, and the running efficiency of the highway is improved. And selecting and executing the most adaptive operation strategy according to the optimization result, thereby further improving the operation efficiency and the service quality of the highway.
Description
Technical Field
The invention relates to the technical field of highway operation and maintenance, in particular to a highway operation method and an information system based on a risk theory.
Background
Traditional highway operation management mode mainly relies on the manual work to carry out information collection and processing, lacks real-time and accuracy. With the progress of technology, various intelligent traffic system applications begin to appear, but most of the intelligent traffic system applications are still in a single-function implementation stage, and comprehensive fusion and deep utilization of information cannot be realized. In addition, when the existing intelligent traffic system is used for solving the complex and sudden road condition problems, the problems of insufficient processing capacity, untimely response and the like still exist. Therefore, a novel highway operation method is urgently needed, various information resources can be better fused and utilized, and highway operation efficiency and service quality are improved.
Disclosure of Invention
The invention aims to provide a highway operation method and an information system based on a risk theory.
In a first aspect, an embodiment of the present invention provides a highway operation method based on a risk theory, which is applied to a highway operation system, where the highway operation system includes a highway network real-time data acquisition interface, a preset operation policy library pre-constructed based on the risk theory, and a visual large screen, and the method includes:
based on a highway network real-time data acquisition interface, acquiring traffic flow information in a current highway network, and identifying highway running health status of the current highway network;
When the abnormal road condition exists in the current road network, carrying out road condition optimization on the current road network according to a current road optimization scheme corresponding to the abnormal road condition;
acquiring a current road network optimization result after road condition optimization, acquiring a current operation strategy aiming at the current road network optimization result from the preset operation strategy library, and analyzing pairing adaptability between the current operation strategy and the current road network;
and executing the current operation strategy on the basis that the pairing adaptability meets preset adaptability conditions, and displaying the execution process of the current operation strategy to the visual large screen.
In a second aspect, an embodiment of the present invention provides an information system, including a server, where the server is configured to perform the method described in the first aspect.
Compared with the prior art, the invention has the beneficial effects that: by adopting the expressway operation method and the expressway operation information system based on the risk theory, disclosed by the invention, the traffic flow information in the current expressway network is acquired through the real-time data acquisition interface of the expressway network, and the expressway operation health state is identified. If abnormal road conditions exist, road condition optimization is carried out according to the current road optimization scheme. The optimized result and the corresponding operation strategy are obtained from a preset operation strategy library, and the pairing adaptability of the optimized result and the current highway network is analyzed. And when the pairing adaptability meets preset conditions, executing the operation strategy, and displaying an execution process on the visual large screen. By the design, the running condition of the highway network can be obtained accurately in real time, abnormal road conditions can be identified and processed rapidly, and the running efficiency of the highway is improved. And selecting and executing the most adaptive operation strategy according to the optimization result, thereby further improving the operation efficiency and the service quality of the highway.
Drawings
In order to more clearly illustrate the technical solution of the embodiments of the present invention, the drawings that are required to be used in the embodiments will be briefly described. It is appreciated that the following drawings depict only certain embodiments of the invention and are therefore not to be considered limiting of its scope. Other relevant drawings may be made by those of ordinary skill in the art without undue burden from these drawings.
Fig. 1 is a schematic step flow diagram of a highway operation method based on risk theory according to an embodiment of the present invention;
fig. 2 is a schematic block diagram of a computer device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
In order to solve the technical problems in the foregoing background, fig. 1 is a schematic step flow diagram of a highway operation method based on a risk theory provided by an embodiment of the present disclosure, where the highway operation method based on the risk theory is applied to a highway operation system, and the highway operation system includes a highway network real-time data acquisition interface, a preset operation policy library pre-constructed based on the risk theory, and a visual large screen, and the highway operation method based on the risk theory is described in detail below.
Step S201, based on a highway network real-time data acquisition interface, acquiring traffic flow information in a current highway network, and identifying highway operation health status of the current highway network;
step S202, when the abnormal road condition exists in the current road network, carrying out road condition optimization on the current road network according to a current road optimization scheme corresponding to the abnormal road condition;
step S203, obtaining a current road network optimization result after road condition optimization, obtaining a current operation strategy aiming at the current road network optimization result from the preset operation strategy library, and analyzing pairing adaptability between the current operation strategy and the current road network;
step S204, executing the current operation strategy on the basis that the pairing adaptability meets preset adaptability conditions, and displaying the execution process of the current operation strategy to the visual large screen.
In the embodiment of the invention, the expressway operation system can collect data in real time through sensors and monitoring equipment arranged at key nodes (such as expressway entrance, expressway exit, expressway service area and the like) of the expressway, and the data comprise traffic flow, vehicle type, vehicle speed, weather conditions, road surface conditions and the like. These data are sent over a wireless network to a central server for processing and analysis. For example, if the average speed of a small vehicle on a certain road suddenly drops, but the weather condition is good and the traffic flow is not increased, the system may determine that the road is in a traffic accident or damaged road, and then mark the road as an abnormal road condition. After the system detects that a traffic accident occurs on a certain section of road, a preset road optimization scheme can be automatically started. For example, the scheme may include immediately dispatching rescue vehicles to the road section, and issuing accident information and detour advice to surrounding drivers through an electronic message board or a mobile phone APP, while also adjusting traffic signal timing of nearby road sections to eliminate congestion as soon as possible. After implementing the optimization scheme, the system will re-collect and analyze road network data to evaluate the optimization effect. For example, the time when the rescue vehicle arrives at the scene is checked to see whether there is a significant drop in the vehicle flow. Meanwhile, the system also selects an operation strategy which is most suitable for the current road condition from a preset operation strategy library, for example, the traffic cost of an accident road section can be increased, so that a driver is encouraged to select other routes. If the matching degree of the current operation strategy and the current highway network state is higher than a preset threshold value, executing the operation strategy is started. For example, the system may instruct to increase the toll on certain road segments and initiate a corresponding toll collection procedure at the toll gate. Meanwhile, all information and operations can be displayed in real time through a large screen, and operators can see a traffic flow chart, an accident handling progress, an adjusted charging standard and the like so as to make further decisions or adjustments. It should be understood that, in the embodiments of the present invention, the risk theory may refer to a highway risk theory, and in the embodiments of the present invention, the highway risk theory mainly focuses on various risks that may occur during the operation of the highway, including but not limited to traffic accidents, bad weather conditions, road damages, etc., and predicts and evaluates the probability of occurrence of these risk events and the possible consequences thereof in combination with probability theory and statistical data.
In one possible embodiment, the aforementioned step S201 may be implemented by the following implementation.
(1) Acquiring traffic flow information in a current highway network, wherein the traffic flow information comprises at least one vehicle running detail;
(2) Performing instant data mining on the vehicle running details to obtain road running state information of the current road network on at least one monitoring index;
(3) Determining expected highway running state information of the current highway network in a preset time range according to the highway running state information through a highway running state evaluation model;
(4) Verifying the expected highway running state information according to the actual highway running state information of the current highway network within the concerned time range obtained by instant data mining to obtain required highway running state information, wherein the preset time range comprises the concerned time range, and the concerned time range corresponds to an evaluation flow of the highway running state evaluation model;
(5) Carrying out highway operation health status identification on the current highway network according to the required highway operation status information;
(6) When the abnormal road condition exists in the current road network, the road condition of the current road network is optimized according to the current road optimization scheme corresponding to the abnormal road condition.
In the embodiment of the invention, the expressway operation system can collect data in real time through the sensors and the monitoring equipment arranged at each key node of the expressway. These data include not only the flow rate of the vehicle but also the driving details of each vehicle, such as the type of vehicle, speed, driving time, lane used, etc. The system analyzes and processes the collected vehicle running details in real time, and identifies key running state indexes of the road network, such as traffic density, average vehicle speed and the like, through an algorithm. The system uses a highway operating condition assessment model to predict highway operating conditions, such as predicted traffic flow, likely congestion segments, etc., over a period of time in the future (e.g., the next hour) based on existing highway operating condition information. When the preset time range is reached (for example, after one hour), the system acquires and analyzes the actual running state information of the road network again, compares the actual running state information with the previously predicted running state of the road, verifies the accuracy of the prediction model, and updates or corrects the running state information of the road according to the actual situation. Based on the verified and corrected road operation status information, the system evaluates the health of the current road network. For example, if the actual traffic flow of a road segment is far beyond expected and this condition persists for a long period of time, the system may mark the road segment as a congested road segment, entering an abnormal road condition. Once the system detects that an abnormal road condition exists in the road network, for example, a certain road section is congested, the system can select a corresponding optimization scheme according to the abnormal road condition. For example, the system may adjust traffic lights on surrounding road segments or issue detour advice on the electronic message board to relieve the pressure on congested road segments.
In one possible implementation manner, the step of verifying the expected highway operation state information according to the actual highway operation state information of the current highway network within the concerned time range obtained by instant data mining to obtain the required highway operation state information may be implemented in the following manner.
(1) Determining the actual road running state information of the current road network within the concerned time range, which is obtained by instant data mining;
(2) Determining an adaptation relation between the actual road running state information and the expected road running state information to verify the expected road running state information;
(3) And when the adaptation relation accords with a preset adaptation relation, determining the expected highway operation state information as required highway operation state information through verification of the expected highway operation state information.
In embodiments of the present invention, the time range of interest may be set to an early peak period (e.g., 7:00-9:00). The expressway operation system collects information such as traffic flow, vehicle type and vehicle speed of each road in two hours in real time through the sensor and the monitoring equipment. For example, the system may find that the traffic flow of a main road continues to increase during these two hours, and that the speed of most vehicles is significantly lower than normal. After the actual highway operation state information is obtained, the system compares the information with the expected highway operation state information of the same time period. For example, if the expected information shows that the traffic flow of the main road increases during the early peak, but the speed should be maintained in the normal range, the decrease in the actual vehicle speed can be considered an abnormal phenomenon, indicating that the actual running state deviates from the expected running state. If the deviation is within a predetermined tolerance (e.g., the vehicle speed is reduced by no more than 10%), the system considers the actual operating condition to be substantially consistent with the expected operating condition, and thus identifies the expected highway operating condition information as the desired highway operating condition information. This means that despite some deviations, the overall operation of the road network is still acceptable, without special treatment or optimization.
In one possible implementation, the present embodiment also provides the following implementation.
(1) And under the condition that the adaptation relation does not accord with a preset adaptation relation, carrying out calibration operation on the expected highway running state information according to the actual highway running state information to obtain the required highway running state information.
In the embodiment of the invention, if the system finds that the deviation between the actual running state and the expected running state exceeds the preset tolerance range, for example, the speed of a certain main road is reduced by more than 10%, the expected road running state information needs to be calibrated. In particular, the system may readjust parameters in the expected model or introduce new variables (such as weather, accidents, etc.) to improve the predictive outcome. In this way, the system can generate a new, more realistic expected highway operation state information, i.e., "demand highway operation state information".
In one possible implementation manner, the step of performing calibration operation on the expected highway operation state information according to the actual highway operation state information to obtain the required highway operation state information may be implemented by the following specific implementation manner,
(1) Determining an actual specific gravity coefficient corresponding to the actual road running state information and an expected specific gravity coefficient corresponding to the expected road running state information;
(2) And according to the actual proportion coefficient and the expected proportion coefficient, performing a combination operation to perform a weighted combination operation to the actual road running state information and the expected road running state information to obtain the required road running state information.
In embodiments of the present invention, the system may take into account a number of factors in determining the specific gravity coefficient. For example, the specific gravity coefficients of actual road operation state information and expected road operation state information may vary in different time periods during the day due to differences in traffic flow and driver behavior patterns. For example, in the early rush hour, the actual specific gravity coefficient of the actual road running state information may be set higher, for example, 0.7, due to the large traffic flow and complicated road conditions; whereas during late night hours the traffic flow is small and the driver behavior pattern is relatively stable, the expected specific gravity factor of the expected road running state information may be set higher, for example 0.6. After the actual specific gravity coefficient and the expected specific gravity coefficient are obtained, the system performs weighted combination on the actual road operation state information and the expected road operation state information. Taking the early peak period as an example, assuming that the actual road operation state information shows 5000 traffic per hour for a certain main road and 4500 traffic per hour for the expected road operation state information, the required road operation state information will be: demand traffic = 0.7 x 5000+0.3 x 4500 = 3500+1350 = 4850 vehicles/hour. This result reflects both the current actual situation and also considers predictions of the expected model.
In one possible implementation, the following examples are also provided in the examples of the present invention.
(1) And training the road running state evaluation model according to the actual road running state information under the condition that the adaptation relation does not accord with the preset adaptation relation, so as to obtain the trained road running state evaluation model.
In the embodiment of the invention, when the system finds that the deviation between the actual running state and the expected running state exceeds the preset tolerance range, for example, the traffic flow of a main road in the early peak period is far beyond the predicted value, the road running state evaluation model needs to be adjusted. Specifically, the system may input current actual running state information (such as traffic flow, speed, etc.) and actual road conditions (such as congestion or not) as new training samples into the road running state evaluation model. Through machine learning algorithms, such as decision trees, neural networks, etc., the model automatically adjusts its internal parameters to more accurately predict future highway operating conditions. After a period of training, a trained, more accurate road running state assessment model is obtained. The model can better reflect the operation rule of the actual road network, thereby improving the prediction accuracy and the management efficiency of the road network.
In one possible implementation manner, the step of training the road running state evaluation model according to the actual road running state information to obtain a trained road running state evaluation model may be implemented in the following manner.
(1) Determining a cost parameter between the actual road running state information and the expected road running state information;
(2) And updating the model parameters of the road running state evaluation model according to the cost parameters to obtain the road running state evaluation model after training.
In an embodiment of the invention, the actual traffic flow of a road is found to be 5000 vehicles per hour, and the expected traffic flow is found to be 4500 vehicles per hour, for example, during the early peak hours of the day. Here, the deviation between the actual and the expected is 500 pieces/hour. If a Mean Square Error (MSE) is used as the cost parameter, the result of the calculation will be the square of the error, i.e., (500)/(2=250000). The larger this value, the worse the predictive effect of the model, requiring more training to reduce the error. It is assumed that a neural network-based road running state evaluation model is used, and the model comprises parameters such as weight, bias and the like. By means of back propagation, gradient descent and other methods, the system can update the parameters according to the calculated cost parameters so as to achieve the aim of minimizing the cost parameters. For example, if the original weight is 0.5, the bias is 100, and after a round of training, the weight may be updated to 0.45, and the bias is updated to 105. Thus, the model can obtain a result which is closer to the actual situation when predicting next time. In this process, it may be necessary to iterate and calibrate the model parameters multiple times, and continuously compare the difference between the new expected road running state information and the actual road running state information until the prediction result of the model reaches satisfactory accuracy, that is, the cost parameter reaches an acceptable range.
In one possible embodiment, the step of determining the adaptation relation between the actual road operation state information and the expected road operation state information may be implemented in the following manner.
(1) Executing time dimension adjustment operation on the data stream corresponding to the expected highway running state information under the condition that the difference between the time spans of the preset time range and the concerned time range is larger than a preset span threshold value, so as to obtain an adjusted expected data stream;
(2) Executing time dimension adjustment operation on the data stream corresponding to the actual road running state information to obtain an adjusted actual data stream;
(3) Determining a correspondence between the adjusted expected data stream and the adjusted actual data stream;
(4) And according to the consistency, determining an adaptation relation between the actual road running state information and the expected road running state information.
In an embodiment of the present invention, if the preset time range is morning (e.g., 7:00-9:00), but the time range of interest is afternoon (e.g., 13:00-15:00), then the expected data flow needs to be adjusted to accommodate this change. Specifically, the system may find trends in vehicle flow and vehicle speed under the same weather, workday/non-workday, etc. conditions during these two time periods by looking at the historical data, and then adjust the expected data flow based on these trends. In order to enable the actual data stream to be compared with the expected data stream, the system also requires an adjustment of the time dimension for the actual data stream. For example, the system may extract the traffic and vehicle speed data for the midday period and then resample and interpolate at intervals of the expected data stream (e.g., one data point every 15 minutes) to obtain a new actual data stream. After the adjustment of the time dimension is completed, the system may use some consistency check method to compare the expected data stream with the actual data stream. For example, if pearson correlation coefficient is used as a measure of consistency, it may be found that in the above scenario, the correlation coefficient of the adjusted expected data stream and the adjusted actual data stream is 0.8, which indicates that both have a high consistency. If the result of the consistency check exceeds a preset threshold (e.g., the correlation coefficient is greater than 0.75), the system considers that there is a good fit between the actual road operation state information and the expected road operation state information, and can be directly used for evaluation and prediction of the road operation state. If the consistency is below the threshold, further optimization and tuning of the expected model may be required to improve its prediction accuracy.
In one possible implementation manner, the step of optimizing the road condition of the current road network according to the current road optimization scheme corresponding to the abnormal road condition may be implemented in the following manner.
(1) Under the condition that the abnormal road condition represents that the current road network has a hidden trouble of road facility damage, an emergency maintenance repair scheme is executed aiming at the current road network so as to optimize the road condition of the current road network;
(2) Under the condition that the abnormal road condition represents that the current road network has traffic jam hidden danger, executing a traffic flow dredging and scheduling scheme aiming at the current road network so as to optimize the road condition of the current road network;
(3) Under the condition that the abnormal road condition characterizes that the current road network has abnormal weather hidden trouble, executing a bad weather response strategy aiming at the current road network so as to optimize the road condition of the current road network;
(4) Under the condition that the abnormal road condition represents that the current road network has unknown vehicle entrance hidden trouble, vehicle feature recognition is carried out on vehicles to be recognized entering the current road network so as to determine vehicle recognition information of the vehicles to be recognized;
(5) And under the condition that the vehicle identification information represents that the vehicle to be identified is an illegally driven vehicle, an illegal vehicle supervision scheme is executed aiming at the current road network so as to optimize the road condition of the current road network.
In the embodiment of the invention, the system is supposed to detect that a certain section of road surface of a main road is seriously damaged, and the normal running of the vehicle can be influenced. In this case, the system may immediately initiate an emergency maintenance repair scheme, such as dispatching an engineering team to repair as soon as possible, while informing the driver to avoid the road surface via the information distribution system. If the system predicts that a road may be jammed in a future period of time, measures are required to clear the traffic. A specific scheduling scheme may include adjusting the timing of traffic lights or recommending other, more unobstructed routes to the driver via the information distribution system. When the system detects that severe weather such as snow may occur in a region and such weather may affect the traffic conditions of the road, then a severe weather response strategy may need to be initiated. This may include making preparations for anti-skid facilities in advance, or closing some snow-prone roads when necessary. If the system finds that an unknown vehicle is driving into the road network, such as a large truck that is not registered in the system, it is necessary to identify the vehicle characteristics, including information about the vehicle type, license plate number, etc. If the system determines that the large truck is unauthorized to be driven into the road network through vehicle feature identification, then an illegal vehicle supervision scheme needs to be initiated. Specific measures may include informing traffic police authorities to handle or alerting drivers through electronic screens, etc.
In another embodiment of the invention, the system may facilitate emergency maintenance repair schemes if a bridge exhibits possible structural problems, such as cracks or damage. Specific operations may include immediately closing the bridge, preventing traffic, scheduling engineering teams for field inspection and necessary maintenance, and informing drivers of the change of course by broadcasting, electronic guideboards, etc. Assuming that the system predicts that a major road may experience severe congestion during the next few hours due to the handling of a large event. At this point, the system may take action in advance, such as modifying the timing of surrounding traffic lights, optimizing traffic flow, guiding portions of the vehicle to use other roads, or suggesting active participants to use public transportation. When a weather system predicts that a snow is imminent and may have an impact on the road network, the system may take various bad weather countermeasures. This may include pre-salting to prevent ice on the road, deploying a snow team ready to remove snow, or alerting the driver to safety through broadcast and electronic guideboards to slow down travel. If the system monitors that an unrecognizable vehicle has entered the highway, it may be because the vehicle has not properly registered or is a stolen vehicle. The system can acquire the characteristic information of the vehicle, such as the vehicle model, the color, the license plate and the like, through a high-definition camera, a radar or other sensors. If, by the vehicle feature identification of the previous step, the system determines that this is a stolen vehicle or a vehicle that is not registered as specified, the system initiates an illegal vehicle supervision scheme. Such as informing police departments, recording their travel routes, restricting them from entering certain areas, and even taking remote control to stop the vehicle.
In one possible embodiment, the foregoing step of performing an illegal vehicle supervision scheme for the current road network may include the following embodiments.
(1) According to a preset illegal driving-in vehicle list, executing a blocking operation on a subsequent passing application of the vehicle to be identified aiming at the current highway network;
(2) According to the vehicle running condition of the vehicle to be identified which enters the current highway network, performing association inquiry operation on vehicle running details corresponding to the vehicle to be identified;
(3) And according to a preset peer-to-peer vehicle screening strategy, performing illegal peer-to-peer vehicle screening operation on the traffic flow information in the current highway network.
In embodiments of the present invention, for example, a database is built into the system, which contains a list of all known illegal vehicles that may be blacklisted for unpaid fines, outstanding annual checks, or other violations. When a vehicle requests to enter a particular area or road, the system automatically queries whether the vehicle is illegally driving into the list of vehicles. If the match is successful, the system will reject the vehicle's application for traffic and may trigger an alarm or notify traffic authorities for further processing. For example, the system may collect and analyze various data including, but not limited to, vehicle speed, direction of travel, braking frequency, etc., as well as information from other sources (e.g., traffic monitoring cameras, GPS trackers, etc.). The system may also query the behavior patterns of the vehicle at a particular time and place based on the travel history of the vehicle. For example, if a vehicle frequently goes out in a certain area late at night, or its speed frequently exceeds a limit, the system may mark it as a suspicious vehicle and conduct further investigation. The system may analyze the driving pattern of the vehicle, especially if there is a cooperative driving of a plurality of vehicles, according to a series of algorithms and models, such as machine learning or pattern recognition techniques. If the system finds that two or more vehicles are always traveling at the same time, place, and similar speed and direction, the system may infer that these vehicles are likely to be moving together. Such information may be used to reveal potential illegal activities such as illegal racing, smuggling, or other criminal activities. At this point, the system will automatically alert the relevant department to the check.
In one possible implementation, the embodiment of the present invention further includes the following steps.
(1) Performing instant hidden danger analysis operation on the road running state information to obtain road hidden danger analysis conditions;
(2) And under the condition that the analysis condition of the hidden danger of the highway represents that the current highway network has abnormal road conditions, carrying out road condition optimization on the current highway network according to the current highway optimization scheme corresponding to the abnormal road conditions.
In the present embodiment, it is assumed that the system is monitoring a highway network of a city. The system may gather various highway operating state information including, but not limited to, traffic flow, vehicle speed, traffic accident recording, weather conditions, and the like. The system then performs real-time hidden danger analysis on the data. For example, if the traffic volume of a certain road is suddenly increased or if a certain traffic node has a plurality of small collision accidents, the system may determine that traffic hazards exist in the places. In the above example, if the system determines that there is a traffic hazard on a road or a traffic node, then action is required to optimize the current highway operating state. Specific optimization schemes may include adjusting the timing of traffic lights or guiding the driver through broadcasting and electronic guideboards to change the route. If the situation is severe, it may also be necessary to temporarily shut down certain roads or traffic nodes to ensure traffic safety.
In one possible embodiment, the invention is also applicable to a road management device comprising a vehicle monitoring unit; the step of acquiring traffic information in the current road network, where the traffic information includes at least one vehicle driving detail, may include the following examples.
(1) Acquiring traffic flow information in a current road network through the vehicle monitoring unit, wherein the traffic flow information comprises at least one vehicle driving detail;
the step of performing the instant data mining on the vehicle driving details to obtain the road running state information of the current road network on at least one monitoring index may include the following examples. (1) The vehicle monitoring unit is used for carrying out instant data mining on the vehicle running details to obtain road running state information of the current road network on at least one monitoring index;
the step of performing the operation of analyzing the real-time hidden danger on the road running state information to obtain the analysis condition of the hidden danger may include the following examples.
(1) And the road running state information is immediately sent to the road hidden danger monitoring equipment through the data transmission interface of the vehicle monitoring unit so as to start the road hidden danger monitoring equipment to execute instant hidden danger analysis operation on the road running state information, and the road hidden danger analysis condition is obtained.
In the embodiment of the present invention, it is assumed that there is one vehicle monitoring unit in a highway network of one city. This unit may contain various sensors and devices such as traffic cameras, speed radars, GPS trackers, etc. The method can collect and record the running details of the vehicle in real time, including the position, speed, direction, running time and the like of the vehicle. After collecting the vehicle driving details, the vehicle monitoring unit processes and analyzes the data in real time. For example, it may calculate the average speed of the vehicle on each road, or count the traffic at a certain intersection. The data form road running state information and can be used for reflecting the actual performance of the road network on various monitoring indexes. After the road running state information is acquired and processed, the vehicle monitoring unit can send the information to the road hidden trouble monitoring device through the data transmission interface. The device may be a server or computer system with advanced data processing capabilities. The method can further analyze hidden danger of the received information, such as detecting whether traffic jam, road damage or other abnormal road conditions exist.
In one possible implementation manner, the step of determining the expected highway operation state information of the current highway network in the preset time range according to the highway operation state information through the highway operation state estimation model may be implemented by the following implementation manner.
(1) Transmitting the highway running state information to highway running state evaluation equipment through a public data channel, wherein a highway running state evaluation model is maintained on the highway running state evaluation equipment;
(2) And starting the highway running state evaluation equipment to determine expected highway running state information of the current highway network in a preset time range according to the highway running state information through a highway running state evaluation model.
In the embodiment of the invention, it is assumed that in a highway network of a city, a vehicle monitoring unit collects a great amount of highway operation state information, including traffic flow, speed, accident record, and the like. This information is then sent to the road running status assessment device via a public data channel, such as the internet or a private network. The device may be a high performance server that maintains a highway operating condition assessment model. This model may be built based on machine learning or other statistical methods to predict the operational state of the road network over a period of time in the future. When the highway operation state assessment device receives the highway operation state information, it starts the highway operation state assessment model, and calculates the information as input. For example, if now an early rush hour, the model may predict that in the next hour, the traffic in a heavy traffic area will increase and the traffic in a residential area will decrease. This expected highway operating state information may be used to help city administrators better schedule traffic resources to address possible traffic problems in advance.
In one possible embodiment, the aforementioned step S203 may be implemented by the following specific embodiment.
(1) Acquiring a current road network optimization result of a current road network and a current operation strategy aiming at the current road network optimization result;
(2) Determining a road network optimization vector of the current road network according to the past optimization data of the current road network, wherein the past optimization data comprises a past road network optimization result of the current road network and a past operation strategy aiming at the past road network optimization result;
(3) And determining a strategy adaptation degree according to the highway network optimization vector of the current highway network, the current highway network optimization result and the current operation strategy, wherein the strategy adaptation degree is used for representing pairing adaptability between the current operation strategy and the current highway network.
In the embodiment of the present invention, it is assumed that a highway network of one city is being managed. The system may perform optimization operations on the road network periodically or on demand, such as adjusting timing of traffic lights, directing vehicles to change driving routes, etc. The result of these operations is a road network optimization result. At the same time, in order to achieve these optimized operations, the system may also take a series of operation strategies, such as broadcasting traffic information, setting up electronic guideboards, etc. After collecting road network optimization results and corresponding operating strategies over a period of time, the system may calculate a road network optimization vector. This vector may contain various indicators, such as average vehicle speed, traffic accident rate, traffic flow, etc., which may reflect the behavior of the road network under different operating strategies. Once with the road network optimization vector, the system can evaluate the pairing adaptability between the current operation strategy and the current road network. This adaptation may be a number or percentage, the higher the current operating policy the more suitable the current road network state. For example, if the policy adaptation is high, it means that the current operation policy is successful in improving the efficiency and security of the road network. Conversely, adjustment of the operation policy may need to be considered.
In another embodiment of the present invention, it is assumed that the following optimization result is obtained after the current road network is optimized by road conditions: the traffic time of the congested road section is reduced, and the overall vehicle traffic speed is improved. And then selecting a corresponding strategy from a preset operation strategy library, such as adjusting a signal lamp timing scheme in advance to optimize the traffic fluency. According to past optimization records, past highway network optimization results are analyzed, and different operation strategies are used for the results. The road network optimization vector of the current road network is obtained through calculation and statistics, and the road network optimization vector comprises indexes such as reduction of road congestion degree, improvement of green wave pass rate, reduction of traffic accident rate and the like. And calculating a strategy adaptation value by using an adaptation evaluation algorithm based on the road network optimization vector, the current road network optimization result and the current operation strategy. For example, the fitness may be determined based on the weights of the various indicators in the optimization vector and the deviation between the current optimization result and the intended target. And if the current operation strategy is highly matched with the optimization result of the current highway network, the adaptation degree is higher. The method is designed in such a way, and a strategy adaptation value is calculated by using an adaptation evaluation algorithm based on the road network optimization vector, the current road network optimization result and the current operation strategy. For example, the fitness may be determined based on the weights of the various indicators in the optimization vector and the deviation between the current optimization result and the intended target. And if the current operation strategy is highly matched with the optimization result of the current highway network, the adaptation degree is higher.
In one possible implementation, the step of determining the road network optimization vector of the current road network according to the past optimization data of the current road network may be implemented by the following implementation.
(1) Determining a plurality of target road network optimization result nodes corresponding to the current road network in a road network optimization semantic network according to the past road network optimization result of the current road network, wherein the road network optimization semantic network is used for representing the correlation among the plurality of road network optimization result nodes;
(2) Determining a plurality of target operation strategy nodes corresponding to the current road network in an operation strategy semantic network according to the past operation strategy of the current road network, wherein the operation strategy semantic network is used for representing the correlation among the plurality of operation strategy nodes;
(3) And determining the road network optimization vector of the current road network according to the target road network optimization result nodes and the target operation strategy nodes.
In the embodiment of the invention, firstly, the system can find the corresponding node in the highway network optimization semantic network according to the past optimization result of the highway network. The semantic network is a graphical data structure that represents the relationships between different road network optimization results. For example, if in past optimization operations, it was found that adjusting traffic signal timing can effectively reduce traffic congestion, then the two nodes "traffic signal timing" and "traffic congestion" may have a strongly correlated edge. Similarly, the system can find corresponding nodes in the operation strategy semantic network according to the past operation strategy. For example, if during past operation it was found that guiding a vehicle to change travel routes during early peak periods could effectively relieve traffic pressure, then the two nodes "change travel routes" and "early peak" may have a strongly correlated edge. Finally, the system calculates a road network optimization vector according to the found road network optimization result node and the operation strategy node. This vector may contain various indicators, such as average vehicle speed, traffic accident rate, traffic flow, etc., which may reflect the behavior of the road network under different operating strategies.
In another embodiment of the present invention, it is assumed that multiple road condition optimizations have been performed on the current road network in the past, and different road network optimization results are obtained. These optimization results reflect different road network conditions, such as reduced congestion, increased vehicle traffic speed, etc. By constructing a highway network optimization semantic network, each optimization result is used as a node, and the relevance between the nodes is established. For past road network optimization results, a different operating strategy is used for each optimization result. These past operating strategies may be adjusting signal timing, limiting traffic flow, or enforcing traffic control, etc. By constructing an operation strategy semantic network, each operation strategy is used as a node, and the relevance between the operation strategies is established. And comprehensively analyzing the obtained target highway network optimization result node and the target operation strategy node to determine the highway network optimization vector of the current highway network. The optimization vector may be a vector comprising a plurality of dimensions, each dimension representing a particular road network indicator or operational policy indicator. For example, traffic flow, transit time, congestion level, signal timing, etc. may be considered as different dimensions and combined to form a comprehensive road network optimization vector. By the design, a highway network optimization semantic network and an operation strategy semantic network can be constructed according to past data, and the highway network optimization vector of the current highway network is determined by utilizing the networks. The method can help understand the relevance between the past optimization result and the operation strategy, and further guide the optimization decision and the operation strategy selection of the current highway network.
In one possible implementation manner, the step of determining the road network optimization vector of the current road network according to the plurality of target road network optimization result nodes and the plurality of target operation policy nodes may be implemented by the following implementation manner.
And executing combination operation on the plurality of optimization result vectors and the plurality of operation strategy vectors to obtain a highway network optimization vector of the current highway network, wherein the optimization result vector is a feature vector corresponding to the target highway network optimization result node, and the operation strategy vector is a feature vector corresponding to the target operation strategy node.
In the embodiment of the invention, after the target road network optimization result node and the target operation strategy node are determined, the system further calculates the feature vectors corresponding to the nodes. The feature vector may be viewed as a mathematical tool that captures the location and importance of the nodes in the semantic network. For example, if a "traffic light" node is closely connected to many other nodes, its eigenvector may be very large. The system then combines the optimization result vector with the operation policy vector. This operation may include vector addition, multiplication, or other mathematical operations. The result is the road network optimization vector of the current road network, integrates the optimization result and the information of the operation strategies, and can comprehensively reflect the performance of the road network under different operation strategies.
In another embodiment of the invention, it is assumed that there are 3 target highway network optimization result nodes (A, B, C) and 2 target operational policy nodes (X, Y). Each target road network optimization result node has a corresponding feature vector, such as a= [0.8,0.6,0.9], b= [0.7,0.5,0.8], and c= [0.9,0.7,0.6]; each target operation policy node also has a corresponding feature vector, such as x= [0.5,0.4], y= [0.6,0.3]. In performing the combining operation, these vectors may be combined using weighted summation or other combining means. And combining the feature vector corresponding to each target road network optimization result node with the feature vector corresponding to each target operation strategy node by executing the combination operation. For example, the feature vector of the target road network optimization result node and the feature vector of the target operation policy node may be added according to a certain weight to obtain the road network optimization vector of the current road network. Assuming that a weighted summation mode is used and the weight given to the road network optimization result is 0.7 and the weight of the operation policy is 0.3, the road network optimization vector of the current road network can be expressed as: [0.7A+0.3X,0.7B+0.3X,0.7C+0.3Y ]. By the design, a plurality of optimization result vectors and a plurality of operation strategy vectors can be combined to obtain the highway network optimization vector of the current highway network. Such vectors represent the comprehensive characteristics of the current road network between different road network optimization results and operation strategies, which are helpful for better understanding and evaluating the optimization status and operation strategy selection of the current road network.
In a possible implementation manner, before the step of performing the combining operation on the plurality of optimization result vectors and the plurality of operation policy vectors to obtain the road network optimization vector of the current road network, the following implementation manner may be further provided.
(1) Executing PageRank algorithm on the highway network optimization semantic network to obtain feature vectors of a plurality of highway network optimization result nodes in the highway network optimization semantic network;
(2) And executing a PageRank algorithm on the operation strategy semantic network to obtain feature vectors of a plurality of operation strategy nodes in the operation strategy semantic network.
In the embodiment of the invention, firstly, the system runs the PageRank algorithm on the highway network optimization semantic network. This algorithm was originally invented by one of the google creator's larly-peter to evaluate the importance of web pages. In this scenario, the PageRank algorithm is used to determine the importance of each road network optimization result node. For example, if a "traffic light timing" node is closely connected to many other nodes, it may be considered very important that its feature vector (i.e., pageRank value) be relatively large. Likewise, the system may also run the PageRank algorithm on the operational policy semantic network to determine the importance of each operational policy node. For example, if a "change route of travel" strategy successfully relieves traffic pressure during past operations, then this strategy node may be considered very important, and its feature vector (i.e., pageRank value) may be relatively large. After the two steps are performed, the system obtains the road network optimization result and the characteristic vector of the operation strategy, which can be used for generating the road network optimization vector.
In another implementation of the embodiment of the invention, a highway network optimization semantic network is assumed, wherein a plurality of highway network optimization result nodes (A, B, C) are included. By executing the PageRank algorithm, the importance or impact of each node can be calculated and a corresponding feature vector generated for each node. For example, after the PageRank algorithm calculation, the feature vector of node A is [0.2,0.4,0.4], the feature vector of node B is [0.3,0.2,0.5], and the feature vector of node C is [0.5,0.3,0.2]. Similarly, it is assumed that there is one operational policy semantic network that contains associations between multiple operational policy nodes (X, Y). By executing the PageRank algorithm, the importance or impact of each operational policy node can be calculated and a corresponding feature vector generated for each node. For example, after the PageRank algorithm calculation, the feature vector of node X is [0.6,0.4], and the feature vector of node Y is [0.4,0.6]. And performing a combination operation on the basis that the feature vectors of a plurality of road network optimization result nodes in the road network optimization semantic network and the feature vectors of a plurality of operation strategy nodes in the operation strategy semantic network are obtained. These vectors may be combined using weighted summation or other means to obtain a road network optimization vector for the current road network. Assuming that a weighted summation mode is used and the weight given to the road network optimization result node is 0.7 and the weight given to the operation policy node is 0.3, the road network optimization vector of the current road network can be expressed as: [0.7A+0.3X,0.7B+0.3X,0.7C+0.3Y ]. By the design, pageRank algorithm can be respectively executed on the highway network optimization semantic network and the operation strategy semantic network to obtain feature vectors of a plurality of highway network optimization result nodes and operation strategy nodes, and then the feature vectors are combined to obtain the highway network optimization vector of the current highway network. Such a method combines relevance analysis and weighted combination, helping to more accurately represent and evaluate the optimal state and operational policy choices of the current road network.
In one possible implementation manner, the step of determining a plurality of target road network optimization result nodes corresponding to the current road network in the road network optimization semantic network according to the past road network optimization result of the current road network may be implemented in the following manner.
(1) Obtaining at least one road network optimization content from a past road network optimization result of the current road network, wherein the road network optimization content comprises association relations between undetermined road network optimization result nodes and other road network optimization result nodes in the past road network optimization result and semantic description contents of the undetermined road network optimization result nodes;
(2) And determining a plurality of target road network optimization result nodes corresponding to the current road network in the road network optimization semantic network according to the road network optimization content.
In the embodiment of the invention, when the optimization problem of the highway network is processed, the system firstly analyzes the past optimization result. For example, if it was found in a previous operation that "increasing road width" could effectively reduce traffic congestion, this information would be recorded. Such information may include other optimization results related to "increase road width" (e.g., "improve traffic signal timing")(s) and semantic description of the operation of "increase road width. After all the optimization content is collected, the system can find the corresponding node in the highway network optimization semantic network. For example, "increasing road width" and "improving traffic signal timing" may be two adjacent nodes, as they are both related to reducing traffic congestion. These nodes become target road network optimization result nodes for subsequent analysis and calculation.
In another implementation manner of the embodiment of the present invention, it is assumed that road condition optimization is performed multiple times on a highway network of a certain city for a period of time. One of the road network optimizations is to reduce traffic congestion and increase road traffic speed. The specific highway network optimization content description may be: the optimization scheme A reduces traffic jam and increases average vehicle speed; the optimization scheme B improves the road traffic capacity and enables vehicles to smoothly pass through. In the road network optimization semantic network, a plurality of road network optimization result nodes exist, and each node represents a different optimization scheme or a specific road condition improvement condition. According to the highway network optimization content, semantic descriptions and association relations can be analyzed, and corresponding target highway network optimization result nodes of the current highway network in the highway network optimization semantic network can be determined. The optimization scheme a may be associated with a past road network optimization result node X, which represents a reduction in traffic congestion and an increase in average vehicle speed. The optimization scheme B may be associated with a past road network optimization result node Y, which represents an improvement in road traffic and vehicle fluency. Thus, from the semantic description and association of the road network optimization content, the target road network optimization result nodes X and Y can be determined, which correspond to the targets of reducing congestion and improving the vehicle passing speed in the current road network. Such a method helps locate target nodes in the road network optimization semantic network that are relevant to the current road network state, providing valuable guidance for further analysis, decision making and improvement.
In a possible implementation manner, the step of determining a plurality of target road network optimization result nodes corresponding to the current road network in the road network optimization semantic network according to the road network optimization content may be implemented in the following manner.
(1) Determining the associated low-dimensional representation and the semantic description low-dimensional representation of the undetermined road network optimization result node according to the road network optimization content;
(2) Acquiring a road network optimization result compression characteristic of the current road network according to the associated low-dimensional representation and the semantic description low-dimensional representation;
(3) And comparing the compressed characteristics of the road network optimization result with the compressed characteristics of a plurality of road network optimization result nodes in the road network optimization semantic network, determining a plurality of target road network optimization result nodes from the plurality of road network optimization result nodes, wherein the characteristic distance between the compressed characteristics of the target road network optimization result nodes and the compressed characteristics of the road network optimization result is smaller than a preset characteristic distance threshold.
In the embodiment of the invention, when processing the road network optimization problem, the system firstly extracts information from the road network optimization content and converts the information into a low-dimensional representation form. For example, if the optimization includes "increasing road width", this information may be converted into a vector containing several digits, which may represent the semantic content of "increasing road width" and the degree of association with other optimization results. The system then computes a compressed feature of the road network optimization result by combining the associated low-dimensional representation with the semantically descriptive low-dimensional representation. This feature is a highly refined data that can effectively represent the core information of the road network optimization results. Finally, the system compares the compressed characteristics of the road network optimization result with the characteristics of the nodes in the road network optimization semantic network. If the distance between the feature of a certain node and the compressed feature of the road network optimization result is smaller than the preset threshold value, the node is regarded as a target road network optimization result node.
In another implementation of the embodiment of the present invention, a highway network optimization semantic network is assumed, which includes the following three highway network optimization result nodes: the node A represents a scheme for adding traffic lights to reduce traffic jams. Node B, which represents a new road extension strategy to increase traffic flow and road capacity. The node C represents an improved scheme of an intelligent traffic signal system, and traffic is optimized by adjusting signal timing in real time. Now, a pending road network optimization result node is determined, which describes a new public transportation route planning scheme. In order to determine its location in the road network optimization semantic network, the following operations may be performed: correlating the low-dimensional representations: and calculating the association degree between the undetermined node and the node A, the node B and the node C by using cosine similarity and other methods. Assuming that the resulting low-dimensional representation vector is [0.8,0.2,0.3], this means that the association of the pending node with node A is higher and the association with node B and node C is lower. Semantic description low-dimensional representation: the textual description of the pending node is converted into a low-dimensional feature vector. For example, a deep learning model (e.g., BERT) is used to convert the text description "provide more public transportation lines, reduce private travel" into a 100-dimensional semantic description low-dimensional representation vector. It is assumed that there is historical road network optimization result data, which includes records of indexes such as traffic flow, speed, congestion index, and the like. From which the compression characteristics of the current road network are extracted. For example, summary information of indexes such as average vehicle speed, road utilization, and traffic congestion index in the past week is calculated. These indices constitute the compressed feature vector of the current road network, assumed as [60,0.7,0.6], where 60 represents the average vehicle speed, 0.7 represents the road utilization, and 0.6 represents the congestion index. The associated low-dimensional representation vector [0.8,0.2,0.3] of the undetermined node is compared with the semantically described low-dimensional representation vector. Meanwhile, the compressed characteristic vector [60,0.7,0.6] of the current road network is compared with the corresponding vector of each road network optimization result node. Assume that euclidean distance is used as the feature distance metric method. First, the Euclidean distance of the associated low-dimensional representation vector between the undetermined node and node A, node B and node C is calculated. Thus, [0,0.6,0.5]. The Euclidean distance between the semantic description low-dimensional representation vector of the undetermined node and the node A, the node B and the node C is then calculated. The distances obtained were assumed to be [1.2,1.0,0.9], respectively. And finally, calculating Euclidean distances between the compressed feature vector of the current road network and the node A, the node B and the node C. The distances obtained were assumed to be [10.5,5.0,7.2], respectively. The distance threshold for compressing the feature vector is 8.0 based on a set threshold, such as a distance threshold of 1.0 for correlating the low-dimensional representation with the low-dimensional representation for semantic description. The target road network optimization result node may be determined to be node C because it is closer to the pending node in relevance, semantic description, and compression characteristics of the current road network. Node a and node B are excluded because their distance exceeds the corresponding threshold. So designed, assume that euclidean distance is used as the feature distance metric method. First, the Euclidean distance of the associated low-dimensional representation vector between the undetermined node and node A, node B and node C is calculated. Thus, [0,0.6,0.5]. The Euclidean distance between the semantic description low-dimensional representation vector of the undetermined node and the node A, the node B and the node C is then calculated. The distances obtained were assumed to be [1.2,1.0,0.9], respectively. And finally, calculating Euclidean distances between the compressed feature vector of the current road network and the node A, the node B and the node C. The distances obtained were assumed to be [10.5,5.0,7.2], respectively. The distance threshold for compressing the feature vector is 8.0 based on a set threshold, such as a distance threshold of 1.0 for correlating the low-dimensional representation with the low-dimensional representation for semantic description. The target road network optimization result node may be determined to be node C because it is closer to the pending node in relevance, semantic description, and compression characteristics of the current road network. Node a and node B are excluded because their distance exceeds the corresponding threshold.
In one possible implementation manner, the step of determining a plurality of target operation policy nodes corresponding to the current road network in the operation policy semantic network according to the past operation policy of the current road network may be implemented by the following implementation manner.
(1) Acquiring at least one operation strategy content from a past operation strategy of the current highway network, wherein the operation strategy content comprises an association relation between a pending operation strategy node and other entities in the past operation strategy and semantic description content of the pending operation strategy node;
(2) And determining a plurality of target operation strategy nodes corresponding to the current road network in the operation strategy semantic network according to the operation strategy content.
In the embodiment of the invention, first, the system analyzes past operation strategies and collects relevant information. For example, if it was found in a previous operation that "guiding the vehicle to change the travel route in the early rush hour" can effectively relieve the traffic pressure, this information is recorded. Such information may include other operational policies related to "change travel route" (e.g., "increase bus shifts") and semantic descriptions of this policy of "change travel route". After all the operation policy contents are collected, the system can find the corresponding node in the operation policy semantic network. For example, "changing the travel route" and "increasing the bus shift" may be two adjacent nodes, as they are both associated with alleviating the early peak traffic pressure. These nodes become target operational policy nodes for subsequent analysis and computation.
In another implementation of the embodiment of the present invention, it is assumed that there is a historical data set in which road network operation strategies over the past year have been recorded, including traffic signal timing, lane planning, speed limiting strategies, and the like. And selecting a pending operation strategy node from the data set for analysis. For example, an operation strategy related to traffic signal timing is selected, i.e. the timing of traffic lights is adjusted to a more rational mode. The target operational policy node is determined using the associated low-dimensional representation and the semantically described low-dimensional representation. By way of example only, and not by way of limitation,
correlating the low-dimensional representations: the selected undetermined operation strategy node (traffic light timing adjustment) is associated with other entities in the past operation strategy, such as traffic lights, vehicles, pedestrians and the like. By calculating the degree of association between the pending node and the entities, an associated low-dimensional representation vector can be obtained. For example, the pending node is found to have a higher degree of association with traffic lights and a lower degree of association with vehicles and pedestrians, resulting in an associated low-dimensional representation vector of [0.8,0.2,0.1]. Correlating the low-dimensional representations: the selected undetermined operation strategy node (traffic light timing adjustment) is associated with other entities in the past operation strategy, such as traffic lights, vehicles, pedestrians and the like. By calculating the degree of association between the pending node and the entities, an associated low-dimensional representation vector can be obtained. For example, the pending node is found to have a higher degree of association with traffic lights and a lower degree of association with vehicles and pedestrians, resulting in an associated low-dimensional representation vector of [0.8,0.2,0.1]. Semantic description low-dimensional representation: and converting semantic description contents of the undetermined operation strategy nodes (traffic light timing adjustment) into low-dimensional feature vectors. The textual description of the operation policy is converted into a fixed-length feature vector using natural language processing techniques, such as Word2Vec or BERT. For example, the description of "optimizing traffic lights, reducing traffic congestion" is converted into a 100-dimensional semantic description low-dimensional representation vector.
In a possible implementation manner, the step of determining a plurality of target operation policy nodes corresponding to the current road network in the operation policy semantic network according to the operation policy content may be implemented in the following manner.
(1) Determining an associated low-dimensional representation and a semantic description low-dimensional representation of the undetermined operation strategy node according to the operation strategy content;
(2) Acquiring the operation strategy compression characteristics of the current highway network according to the associated low-dimensional representation and the semantic description low-dimensional representation;
(3) And comparing the operation strategy compression characteristics with compression characteristics of a plurality of operation strategy nodes in the operation strategy semantic network, determining a plurality of target operation strategy nodes from the plurality of operation strategy nodes, wherein the characteristic distance between the compression characteristics of the target operation strategy nodes and the operation strategy compression characteristics is smaller than a preset characteristic distance threshold.
In the embodiment of the invention, when the road network optimization problem is processed, the system firstly extracts information from the operation strategy content and converts the information into a low-dimensional representation form. For example, if the operational strategy content includes "early rush hour lead vehicle change route", this information may be converted into a vector containing several digits, which may represent the semantic content of "change route" and the degree of association with other strategies. The system then computes a compressed feature of the operating policy by combining the associated low-dimensional representation with the semantically descriptive low-dimensional representation. This feature is a highly refined data that can effectively represent the core information of the operating strategy. Finally, the system compares the operational policy compression characteristics with node characteristics in the operational policy semantic network. A node is considered a target operational policy node if its characteristics are less than a predetermined threshold distance from the operational policy compression characteristics.
In another implementation of the embodiment of the present invention, based on the foregoing scenario, a computing method has been mentioned that correlates the low-dimensional representation with the semantically descriptive low-dimensional representation. In this step, the association relation and semantic description content of the selected undetermined operation policy node (traffic light timing adjustment) are converted into corresponding low-dimensional representation vectors. The operational policy compression characteristics of the current road network may be generated by combining the associated low-dimensional representation with the semantically descriptive low-dimensional representation. Specifically, the associated low-dimensional representation vector and the semantically described low-dimensional representation vector can be fused or connected to obtain a comprehensive operation strategy compression feature vector, and the vector represents the operation strategy information of the current highway network. A feature distance metric method (e.g., euclidean distance) may be used to compare the compressed features of the pending operational policy nodes with the compressed features of other nodes in the operational policy semantic network. By calculating the feature distance, the target operation policy node most similar to the undetermined node can be determined. Assume that there are two target operation policy nodes, node a and node B, with compressed feature vectors [0.6,0.5,0.4] and [0.8,0.7,0.9], respectively. And comparing the operation strategy compression characteristic vector of the undetermined node with the compression characteristic vectors of the node A and the node B. The calculated feature distances are assumed to be 2.3 and 1.8, respectively. If the preset feature distance threshold is set to 2.0, the target operation policy node may be determined to be the node B because the feature distance between the compressed feature of the target operation policy node and the node to be determined is smaller than the preset threshold.
In one possible implementation manner, the step of determining the policy fitness according to the road network optimization vector of the current road network, the current road network optimization result and the current operation policy may be performed by the following steps.
(1) Obtaining the association degree of the optimization result between the road network optimization vector of the current road network and the current road network optimization result;
(2) Acquiring the operation strategy association degree between the road network optimization vector of the current road network and the current operation strategy;
(3) Performing a combination operation on the current road network optimization result and the current operation strategy according to the optimization result association degree and the operation strategy association degree to obtain a fusion optimization vector of the current road network;
(4) And determining the strategy adaptation degree according to the road network optimization vector of the current road network and the fusion optimization vector of the current road network.
In the embodiment of the invention, the system firstly calculates the association degree between the road network optimization vector and the current road network optimization result. For example, if the road network optimization vector contains an indicator of "traffic flow", and the current optimization results show that traffic congestion is effectively reduced by "increasing road width", the correlation between the two may be relatively high. The system also calculates the degree of association between the road network optimization vector and the current operating strategy. For example, if the current operating strategy is "leading vehicles to change routes during early rush hour", and some of the indicators in the road network optimization vector (e.g., "average vehicle speed") improve significantly after executing this strategy, then the correlation between the two may be higher. And then, the system combines the optimization result of the current road network with the current operation strategy according to the correlation degree of the optimization result and the correlation degree of the operation strategy to obtain a fusion optimization vector. The vector integrates the information of the optimization result and the operation strategy, and can more comprehensively reflect the performance of the road network after implementing the specific strategy. Finally, the system compares the road network optimization vector with the fusion optimization vector to determine the fitness of the strategy. If the two vectors are very close, then the current operating strategy can be said to be very suitable for the current road network state, and the strategy adaptation is high.
In another implementation of the embodiment of the invention, it is assumed that the traffic road network in a city is being optimized, and that there is the following data and information: road network optimization vector for current road network: this is a vector of values that contains the optimization parameters for each road network element. For example, parameters such as lane width, speed limit, turning radius, etc. of each road segment may be converted into a multidimensional vector representing the optimization vector of the current road network. Road network optimization results for current road network: this is obtained by simulation or other methods, and describes the optimization results of traffic flow, congestion degree and the like of the current road network under specific conditions. For example, data such as average vehicle speed, vehicle density and the like of each road section can be obtained to form an optimization result of the current road network. Current operation policy: this refers to currently implemented traffic management strategies such as traffic light timing schemes, vehicle traffic restrictions, etc. The current operating policy may be converted into a vector representing its characteristics.
Assume that the correlation coefficient between the optimization vector and the optimization result of the current road network is calculated, and a correlation value of 0.85 is obtained. This means that the current optimization vector has a higher relevance to the optimization result. Assuming that the similarity between the current optimization vector and the operation policy is calculated, a relevance value of 0.75 is obtained. This means that there is a certain correlation between the current optimization vector and the operation policy. And taking the relevance of the optimized result and the relevance of the operation strategy into consideration, and fusing the optimized vector and the operation strategy vector of the current highway network by using a weighted average method. If the correlation degree of the optimization result is given a weight of 0.6, and the correlation degree of the operation strategy is given a weight of 0.4, the fusion optimization vector can be calculated as follows: fusion optimization vector = 0.6 x road network optimization vector of current road network +0.4 x current operation policy. Similarity measures (e.g., cosine similarity) may be used to calculate the degree of similarity between the optimized vector and the fused optimized vector for the current road network. And assuming that the calculated fitness value is 0.92, the optimization result of the current road network and the fusion optimization vector have higher fitness.
In one possible implementation manner, the step of obtaining the association degree of the optimization result between the road network optimization vector of the current road network and the current road network optimization result may be implemented in the following manner.
(1) Determining at least one standard road network optimization result node corresponding to the current road network in a road network optimization semantic network according to the current road network optimization result;
(2) And obtaining the association degree of the optimized result according to the inner product between the road network optimized vector of the current road network and the standard optimized result vector, wherein the standard optimized result vector is a characteristic vector corresponding to the standard road network optimized result node.
In the embodiment of the invention, firstly, the system can find the corresponding node in the highway network optimization semantic network according to the optimization result of the current highway network. For example, if the optimization result of the current road network shows that traffic congestion is successfully reduced by "increasing road width", then in the road network optimization semantic network, the node "increasing road width" may be selected as the standard road network optimization result node. The system then calculates the inner product between the road network optimization vector of the current road network and the standard optimization result vector, which can be regarded as the similarity or association between the two vectors. For example, if a certain element value of the current road network optimization vector represents "traffic flow", and a corresponding element value of the standard optimization result vector represents the effect of "increasing road width" on "traffic flow", then the inner product of the two vectors may reflect the degree of association between the current road network optimization result and the standard optimization result.
In another implementation manner of the embodiment of the invention, the assumed highway network optimization semantic network comprises three standard optimization result nodes which are respectively efficient traffic, balanced traffic and low-carbon and environment-friendly. Based on the optimization results (e.g., higher average vehicle speed, moderate traffic flow) of the current road network, it may be determined that the current road network is most relevant to the high-efficiency transit node. And carrying out inner product operation on the road network optimization vector using the current road network and the characteristic vector of the high-efficiency traffic standard optimization result node. Assuming that the inner product result is 0.86, the optimization vector of the current highway network has higher similarity with the standard optimization result vector of the high-efficiency traffic. The feature vector of the efficient transit node may include attributes such as faster average speed, lower congestion index, etc. The feature vector is used as a standard optimization result vector for carrying out inner product calculation with the highway network optimization vector of the current highway network.
The embodiment of the invention provides a determining machine device 100, wherein the determining machine device 100 comprises a processor and a nonvolatile memory which is maintained with determining machine instructions, and when the determining machine instructions are executed by the processor, the determining machine device 100 executes the expressway operation method based on the risk theory. As shown in fig. 2, fig. 2 is a block diagram of a configuration of a determiner device 100 according to an embodiment of the present invention. The determiner device 100 comprises a memory 111, a processor 112 and a communication unit 113. For data transmission or interaction, the memory 111, the processor 112 and the communication unit 113 are electrically connected to each other directly or indirectly. For example, the elements may be electrically connected to each other via one or more communication buses or signal lines.
The foregoing description, for purpose of explanation, has been presented with reference to particular embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the disclosure to the precise forms disclosed. Many modifications and variations are possible in light of the above teaching.
Claims (10)
1. The highway operation method based on the risk theory is characterized by being applied to a highway operation system, wherein the highway operation system comprises a highway network real-time data acquisition interface, a preset operation strategy library pre-constructed based on the risk theory and a visual large screen, and the method comprises the following steps:
based on a highway network real-time data acquisition interface, acquiring traffic flow information in a current highway network, and identifying highway running health status of the current highway network;
when the abnormal road condition exists in the current road network, carrying out road condition optimization on the current road network according to a current road optimization scheme corresponding to the abnormal road condition;
acquiring a current road network optimization result after road condition optimization, acquiring a current operation strategy aiming at the current road network optimization result from the preset operation strategy library, and analyzing pairing adaptability between the current operation strategy and the current road network;
And executing the current operation strategy on the basis that the pairing adaptability meets preset adaptability conditions, and displaying the execution process of the current operation strategy to the visual large screen.
2. The method of claim 1, wherein the obtaining traffic information in a current road network, and the identifying the road running health status of the current road network, comprises:
acquiring traffic flow information in a current highway network, wherein the traffic flow information comprises at least one vehicle running detail;
performing instant data mining on the vehicle running details to obtain road running state information of the current road network on at least one monitoring index;
determining expected highway running state information of the current highway network in a preset time range according to the highway running state information through a highway running state evaluation model;
determining the actual road running state information of the current road network within the concerned time range, which is obtained by instant data mining;
executing time dimension adjustment operation on the data stream corresponding to the expected highway running state information under the condition that the difference between the time spans of the preset time range and the concerned time range is larger than a preset span threshold value, so as to obtain an adjusted expected data stream;
Executing time dimension adjustment operation on the data stream corresponding to the actual road running state information to obtain an adjusted actual data stream;
determining a correspondence between the adjusted expected data stream and the adjusted actual data stream;
according to the consistency, determining an adaptation relation between the actual highway operation state information and the expected highway operation state information so as to verify the expected highway operation state information;
when the adaptation relation accords with a preset adaptation relation, determining the expected highway operation state information as required highway operation state information through verification of the expected highway operation state information, wherein the preset time range comprises the concerned time range, and the concerned time range corresponds to an evaluation flow of the highway operation state evaluation model;
carrying out highway operation health status identification on the current highway network according to the required highway operation status information;
when the abnormal road condition exists in the current road network, the road condition of the current road network is optimized according to the current road optimization scheme corresponding to the abnormal road condition.
3. The method according to claim 2, wherein the method further comprises:
Under the condition that the adaptation relation does not accord with a preset adaptation relation, determining an actual specific gravity coefficient corresponding to the actual highway running state information and an expected specific gravity coefficient corresponding to the expected highway running state information;
according to the actual proportion coefficient and the expected proportion coefficient, carrying out a combination operation on the actual highway running state information and the expected highway running state information to carry out a weighted combination operation so as to obtain required highway running state information;
the method further comprises the steps of:
under the condition that the adaptation relation does not accord with a preset adaptation relation, determining a cost parameter between the actual road running state information and the expected road running state information;
and updating the model parameters of the road running state evaluation model according to the cost parameters to obtain the road running state evaluation model after training.
4. The method according to claim 2, wherein the optimizing the road condition of the current road network according to the current road optimization scheme corresponding to the abnormal road condition comprises:
under the condition that the abnormal road condition represents that the current road network has a hidden trouble of road facility damage, an emergency maintenance repair scheme is executed aiming at the current road network so as to optimize the road condition of the current road network;
Under the condition that the abnormal road condition represents that the current road network has traffic jam hidden danger, executing a traffic flow dredging and scheduling scheme aiming at the current road network so as to optimize the road condition of the current road network;
under the condition that the abnormal road condition characterizes that the current road network has abnormal weather hidden trouble, executing a bad weather response strategy aiming at the current road network so as to optimize the road condition of the current road network;
under the condition that the abnormal road condition represents that the current road network has unknown vehicle entrance hidden trouble, vehicle feature recognition is carried out on vehicles to be recognized entering the current road network so as to determine vehicle recognition information of the vehicles to be recognized;
under the condition that the vehicle identification information represents that the vehicle to be identified is an illegally driven vehicle, executing a blocking operation on a subsequent passing application of the vehicle to be identified aiming at the current highway network according to a preset illegally driven vehicle list;
according to the vehicle running condition of the vehicle to be identified which enters the current highway network, performing association inquiry operation on vehicle running details corresponding to the vehicle to be identified;
and according to a preset peer-to-peer vehicle screening strategy, illegal peer-to-peer vehicle screening operation is carried out on the traffic flow information in the current road network, so that road condition optimization is carried out on the current road network.
5. The method according to claim 2, wherein the method further comprises:
performing instant hidden danger analysis operation on the road running state information to obtain road hidden danger analysis conditions;
and under the condition that the analysis condition of the hidden danger of the highway represents that the current highway network has abnormal road conditions, carrying out road condition optimization on the current highway network according to the current highway optimization scheme corresponding to the abnormal road conditions.
6. The method of claim 5, adapted to a highway management device comprising a vehicle monitoring unit; the obtaining the traffic information in the current road network, wherein the traffic information comprises at least one vehicle driving detail, and the method comprises the following steps:
acquiring traffic flow information in a current road network through the vehicle monitoring unit, wherein the traffic flow information comprises at least one vehicle driving detail;
the step of performing instant data mining on the vehicle running details to obtain road running state information of the current road network on at least one monitoring index, including:
the vehicle monitoring unit is used for carrying out instant data mining on the vehicle running details to obtain road running state information of the current road network on at least one monitoring index;
The step of performing the operation of analyzing the real-time hidden danger on the road running state information to obtain the analysis condition of the hidden danger of the road comprises the following steps:
and the road running state information is immediately sent to the road hidden danger monitoring equipment through the data transmission interface of the vehicle monitoring unit so as to start the road hidden danger monitoring equipment to execute instant hidden danger analysis operation on the road running state information, and the road hidden danger analysis condition is obtained.
7. The method according to claim 2, wherein determining expected highway operation state information of the current highway network in a preset time range according to the highway operation state information through a highway operation state evaluation model comprises:
transmitting the highway running state information to highway running state evaluation equipment through a public data channel, wherein a highway running state evaluation model is maintained on the highway running state evaluation equipment;
and starting the highway running state evaluation equipment to determine expected highway running state information of the current highway network in a preset time range according to the highway running state information through a highway running state evaluation model.
8. The method of claim 1, wherein the analyzing the pairing suitability between the current operation policy and the current road network comprises:
Acquiring a current road network optimization result of a current road network and a current operation strategy aiming at the current road network optimization result;
obtaining at least one road network optimization content from a past road network optimization result of the current road network, wherein the road network optimization content comprises association relations between undetermined road network optimization result nodes and other road network optimization result nodes in the past road network optimization result and semantic description contents of the undetermined road network optimization result nodes;
determining the associated low-dimensional representation and the semantic description low-dimensional representation of the undetermined road network optimization result node according to the road network optimization content;
acquiring a road network optimization result compression characteristic of the current road network according to the associated low-dimensional representation and the semantic description low-dimensional representation;
performing a comparison operation on the compressed characteristics of the road network optimization result and the compressed characteristics of a plurality of road network optimization result nodes in the road network optimization semantic network, determining a plurality of target road network optimization result nodes from the plurality of road network optimization result nodes, wherein the characteristic distance between the compressed characteristics of the target road network optimization result nodes and the compressed characteristics of the road network optimization result is smaller than a preset characteristic distance threshold value, and the road network optimization semantic network is used for representing the correlation among the plurality of road network optimization result nodes;
Acquiring at least one operation strategy content from a past operation strategy of the current highway network, wherein the operation strategy content comprises an association relation between a pending operation strategy node and other entities in the past operation strategy and semantic description content of the pending operation strategy node;
determining an associated low-dimensional representation and a semantic description low-dimensional representation of the undetermined operation strategy node according to the operation strategy content;
acquiring the operation strategy compression characteristics of the current highway network according to the associated low-dimensional representation and the semantic description low-dimensional representation;
performing a comparison operation on the operation strategy compression characteristic and compression characteristics of a plurality of operation strategy nodes in the operation strategy semantic network, determining a plurality of target operation strategy nodes from the plurality of operation strategy nodes, wherein the characteristic distance between the compression characteristics of the target operation strategy nodes and the operation strategy compression characteristics is smaller than a preset characteristic distance threshold, and the operation strategy semantic network is used for representing correlation among the plurality of operation strategy nodes;
performing a combination operation on a plurality of optimization result vectors and a plurality of operation strategy vectors to obtain a highway network optimization vector of the current highway network, wherein the optimization result vector is a feature vector corresponding to the target highway network optimization result node, and the operation strategy vector is a feature vector corresponding to the target operation strategy node;
Determining at least one standard road network optimization result node corresponding to the current road network in a road network optimization semantic network according to the current road network optimization result;
obtaining the association degree of the optimized result according to the inner product between the road network optimized vector of the current road network and the standard optimized result vector, wherein the standard optimized result vector is a feature vector corresponding to the standard road network optimized result node;
acquiring the operation strategy association degree between the road network optimization vector of the current road network and the current operation strategy;
performing a combination operation on the current road network optimization result and the current operation strategy according to the optimization result association degree and the operation strategy association degree to obtain a fusion optimization vector of the current road network;
and determining policy fitness according to the road network optimization vector of the current road network and the fusion optimization vector of the current road network, wherein the policy fitness is used for representing pairing adaptability between the current operation policy and the current road network.
9. The method of claim 8, wherein before performing a combining operation on the plurality of optimization result vectors and the plurality of operation policy vectors to obtain the road network optimization vector of the current road network, the method further comprises:
Executing PageRank algorithm on the highway network optimization semantic network to obtain feature vectors of a plurality of highway network optimization result nodes in the highway network optimization semantic network;
and executing a PageRank algorithm on the operation strategy semantic network to obtain feature vectors of a plurality of operation strategy nodes in the operation strategy semantic network.
10. An information system comprising a server for performing the method of any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311441433.6A CN117475628B (en) | 2023-11-01 | 2023-11-01 | Expressway operation method and information system based on risk theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311441433.6A CN117475628B (en) | 2023-11-01 | 2023-11-01 | Expressway operation method and information system based on risk theory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117475628A true CN117475628A (en) | 2024-01-30 |
| CN117475628B CN117475628B (en) | 2024-05-03 |
Family
ID=89625163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311441433.6A Active CN117475628B (en) | 2023-11-01 | 2023-11-01 | Expressway operation method and information system based on risk theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117475628B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101261768A (en) * | 2007-03-23 | 2008-09-10 | 天津市国腾公路咨询监理有限公司 | Traffic survey data collection and analysis application system for road network and its working method |
| CN114220259A (en) * | 2021-08-13 | 2022-03-22 | 苏交科集团股份有限公司 | Expressway emergency control method based on data fusion |
| CN115358551A (en) * | 2022-08-08 | 2022-11-18 | 北京掌行通信息技术有限公司 | Expressway drainage analysis method and device, storage medium and terminal |
| CN115565365A (en) * | 2022-08-31 | 2023-01-03 | 阿里巴巴（中国）有限公司 | Expressway data processing method, device, equipment and computer program product |
| US11587436B1 (en) * | 2022-03-29 | 2023-02-21 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods for controlling traffic scheduling strategies in smart cities and Internet of Things (IoT) systems thereof |
| CN116308153A (en) * | 2023-03-01 | 2023-06-23 | 北京图安世纪科技股份有限公司 | Holographic intersection management system and method based on digital twinning |
| CN116434523A (en) * | 2022-11-28 | 2023-07-14 | 长沙理工大学 | Vehicle active safety control method and device based on constraint degree in information perception scene |
| CN116597649A (en) * | 2023-06-09 | 2023-08-15 | 山西省智慧交通研究院有限公司 | Road network traffic flow deduction method based on expressway charging system |
| US20230267829A1 (en) * | 2022-02-23 | 2023-08-24 | Southeast University | Multi-objective optimization control method and system for cooperative ramp merging of connected vehicles on highway |
-
2023
- 2023-11-01 CN CN202311441433.6A patent/CN117475628B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101261768A (en) * | 2007-03-23 | 2008-09-10 | 天津市国腾公路咨询监理有限公司 | Traffic survey data collection and analysis application system for road network and its working method |
| CN114220259A (en) * | 2021-08-13 | 2022-03-22 | 苏交科集团股份有限公司 | Expressway emergency control method based on data fusion |
| US20230267829A1 (en) * | 2022-02-23 | 2023-08-24 | Southeast University | Multi-objective optimization control method and system for cooperative ramp merging of connected vehicles on highway |
| US11587436B1 (en) * | 2022-03-29 | 2023-02-21 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods for controlling traffic scheduling strategies in smart cities and Internet of Things (IoT) systems thereof |
| CN115358551A (en) * | 2022-08-08 | 2022-11-18 | 北京掌行通信息技术有限公司 | Expressway drainage analysis method and device, storage medium and terminal |
| CN115565365A (en) * | 2022-08-31 | 2023-01-03 | 阿里巴巴（中国）有限公司 | Expressway data processing method, device, equipment and computer program product |
| CN116434523A (en) * | 2022-11-28 | 2023-07-14 | 长沙理工大学 | Vehicle active safety control method and device based on constraint degree in information perception scene |
| CN116308153A (en) * | 2023-03-01 | 2023-06-23 | 北京图安世纪科技股份有限公司 | Holographic intersection management system and method based on digital twinning |
| CN116597649A (en) * | 2023-06-09 | 2023-08-15 | 山西省智慧交通研究院有限公司 | Road network traffic flow deduction method based on expressway charging system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117475628B (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10902719B2 (en) | Method of predicting traffic congestion and controlling traffic signals based on deep learning and server for performing the same | |
| US9111442B2 (en) | Estimating incident duration | |
| US9008954B2 (en) | Predicting impact of a traffic incident on a road network | |
| Mahmassain et al. | Evaluation of incident detection methodologies | |
| CN109598930B (en) | Automatic detect overhead closed system | |
| CN116504076A (en) | Expressway traffic flow prediction method based on ETC portal data | |
| CN113276874B (en) | Vehicle driving track processing method and related device | |
| CN112462774A (en) | Urban road supervision method and system based on unmanned aerial vehicle navigation following and readable storage medium | |
| Choi et al. | Proactive strategy for variable speed limit operations on freeways under foggy weather conditions | |
| CN112382097A (en) | Urban road supervision method and system based on dynamic traffic flow and readable storage medium | |
| KR100865939B1 (en) | Traffic information guiding system and method | |
| Vanajakshi | Estimation and prediction of travel time from loop detector data for intelligent transportation systems applications | |
| CN117354347B (en) | Vehicle-road cooperative control method and device for multifunctional intelligent lamp post of Internet of things | |
| Abdel-Aty et al. | Real-time big data analytics and proactive traffic safety management visualization system | |
| EP3774478A1 (en) | Vehicular motion assessment method | |
| CN117475628B (en) | Expressway operation method and information system based on risk theory | |
| CN117496726A (en) | Traffic control coordination method and system based on intelligent traffic system | |
| Louah et al. | Traffic operations at an entrance ramp of a suburban freeway first results | |
| CN116863723A (en) | Use method of digital twin base | |
| Oza et al. | Secure traffic lights: Replay attack detection for model-based smart traffic controllers | |
| Nathanail et al. | Traffic volume responsive incident detection | |
| Golze et al. | Impact analysis of accidents on the traffic flow based on massive floating car data | |
| Karnati et al. | Data Mining Algorithms for Traffic Interruption Detection. | |
| Hayashi et al. | Prioritization of Lane-based Traffic Jam Detection for Automotive Navigation System utilizing Suddenness Index Calculation Method for Aggregated Values | |
| CN117894181B (en) | Global traffic abnormal condition integrated monitoring method and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |