Road network evaluation and road implementation priority ranking method based on multi-source data fusion

Abstract

The invention relates to the field of intelligent transportation, and particularly discloses a road network evaluation and road implementation priority ranking method based on multi-source data fusion. According to the method, the multi-source data is collected and preprocessed to serve as the road network evaluation basis, the running state and potential problems of the road network can be comprehensively reflected, compared with a traditional single data source evaluation method, higher accuracy and reliability are achieved, then the road network is evaluated through the road network comprehensive evaluation index model, and the road network evaluation efficiency is improved. According to the method, the road networks can be quantitatively compared, then the evaluation result of each road network is quantitatively compared through a priority ranking algorithm, the road network which is most required to be adjusted is determined, it is ensured that limited traffic construction funds and resources are preferentially used for the road section which is most required to be improved, and the efficiency and social benefits of road construction and management are improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation technology, and in particular relates to a road network evaluation and road implementation priority sorting method based on multi-source data fusion. Background Art

[0002] The construction and management of road networks have become key issues in urban traffic planning. If the road design and management are reasonable, they can effectively undertake urban transportation tasks. If the road design or management is unreasonable, it will not only fail to undertake urban transportation tasks, but will also have an adverse impact on the city’s social economy.

[0003] At present, traditional road network assessment methods mostly rely on a single data source, such as traffic flow, road conditions or road quality, which cannot fully reflect the actual operating conditions of the road network. Moreover, a single data source has certain limitations when dealing with complex road networks and dynamic traffic environments, resulting in low efficiency in road optimization and resource allocation, which is not conducive to operators to carry out targeted optimization of roads based on the assessment results. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a road network evaluation and road implementation priority sorting method based on multi-source data fusion, so as to solve the problem that the road network cannot be effectively evaluated based on a single data source and cannot be used as a basis for road optimization.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] Road network assessment and road implementation priority sorting method based on multi-source data fusion, including:

[0007] Establish a road network multi-source data collection system to collect multi-source data of roads in the road network;

[0008] Preprocess the collected multi-source data, remove noise data, and perform standardization to obtain standardized multi-source data;

[0009] Integrate various standardized multi-source data to form a comprehensive evaluation index model for the road network;

[0010] Using the road network comprehensive evaluation index model to evaluate each road, and obtaining an evaluation result for each road;

[0011] Using a priority sorting algorithm to sort the evaluation results of each road, and determine the implementation priority of the road;

[0012] The evaluation and ranking results are displayed through a visualization platform for reference by decision makers.

[0013] Preferably, the multi-source data collection method includes radar frequency modulation microwave detection, video detection and vehicle-mounted navigation and positioning system detection.

[0014] Preferably, the pretreatment comprises:

[0015] Use threshold method or confidence distance detection method to identify abnormal data;

[0016] Moving average method, time series method and correlation analysis method were used to fill missing data;

[0017] The deletion method is used to deal with invalid data that has no use value.

[0018] Preferably, the standardized multi-source data includes:

[0019] Traffic flow refers to the number of motor vehicles passing through a certain section of road in a unit time;

[0020] Saturation, which represents the ratio of the actual arriving traffic flow of a lane to the lane’s capacity;

[0021] Queue length refers to the length of the queue of vehicles at the entrance of the intersection, which is divided into the maximum length of the queue of each lane at the entrance of the intersection l max and the average queue length of each lane at the intersection entrance, l;

[0022] Occupancy rate, including spatial occupancy rate Rs and temporal occupancy rate Rt. The spatial occupancy rate indicates the ratio of the total area occupied by vehicles in a specific area of ​​the road to the area of ​​the area. The temporal occupancy rate indicates the ratio of the time occupied by vehicles at a known point on the road to the time interval in a specified time interval.

[0023] Delay time refers to the loss of running time caused by the obstruction of vehicle movement;

[0024] Travel time, which represents the time required for a vehicle to travel a specific road segment;

[0025] Vehicle speed, including the instantaneous speed of a vehicle passing through a certain road section and the average speed of a vehicle passing through a certain area;

[0026] Traffic flow density refers to the number of vehicles per unit length on a lane at a certain point in time;

[0027] The secondary queue rate refers to the ratio of the number of vehicles that did not pass through the intersection in the first signal cycle and entered the secondary queue to the number of all vehicles participating in the queue in the first signal cycle;

[0028] Headway time refers to the time difference when the front ends of two adjacent vehicles pass through the same section.

[0029] Preferably, saturation, headway, queue length, delay time and secondary queue rate are selected as characteristic parameters, and an intersection evaluation index model is established through a BP neural network. Traffic flow, travel time, occupancy, traffic flow density and vehicle speed are selected as characteristic parameters, and a road section evaluation index model is established through a SVM classifier. The intersection evaluation index model and the road section evaluation index model are combined to establish the road network comprehensive evaluation index model.

[0030] Preferably, the importance of roads, service demands and socio-economic impacts are introduced as characteristic parameters, and the evaluation results of roads by the road network comprehensive evaluation index model are combined to determine the implementation priority ranking table of roads.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention collects and pre-processes multi-source data and uses it as the basis for road network evaluation, which can comprehensively reflect the operating status and potential problems of the road network. Compared with the traditional single data source evaluation method, it has higher accuracy and reliability. The road network is then evaluated through a road network comprehensive evaluation index model, so that quantitative comparison can be made between each road network. The evaluation results of each road network are then quantitatively compared through a priority sorting algorithm to determine the road network that most needs adjustment, ensuring that limited transportation construction funds and resources are used first for the sections that most need improvement, thereby improving the efficiency and social benefits of road construction and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A block diagram of the method of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] Embodiment 1:

[0036] See also Figure 1 As shown in the figure, the road network evaluation and road implementation priority sorting method based on multi-source data fusion includes:

[0037] Establish a road network multi-source data collection system to collect multi-source data of roads in the road network;

[0038] Preprocess the collected multi-source data, remove noise data, and perform standardization to obtain standardized multi-source data;

[0039] Integrate various standardized multi-source data to form a comprehensive evaluation index model for the road network;

[0040] Using the road network comprehensive evaluation index model to evaluate each road, and obtaining an evaluation result for each road;

[0041] Using a priority sorting algorithm to sort the evaluation results of each road, and determine the implementation priority of the road;

[0042] The evaluation and ranking results are displayed through a visualization platform for reference by decision makers.

[0043] The multi-source data collection method includes radar frequency modulation microwave detection, video detection and vehicle-mounted navigation and positioning system detection.

[0044] From the above, it can be seen that by collecting and preprocessing multi-source data and using it as the basis for road network evaluation, it can comprehensively reflect the operating status and potential problems of the road network. Compared with the traditional single data source evaluation method, it has higher accuracy and reliability. Then, the road network is evaluated through the road network comprehensive evaluation index model, so that each road network can be quantitatively compared. Then, the evaluation results of each road network are quantitatively compared through the priority sorting algorithm to determine the road network that most needs adjustment, ensuring that limited transportation construction funds and resources are used first for the sections that most need improvement, thereby improving the efficiency and social benefits of road construction and management.

[0045] The pre-processing comprises:

[0046] Use threshold method or confidence distance detection method to identify abnormal data;

[0047] Moving average method, time series method and correlation analysis method were used to fill missing data;

[0048] The deletion method is used to deal with invalid data that has no use value.

[0049] The standardized multi-source data includes:

[0050] Traffic flow refers to the number of motor vehicles passing through a certain section of road in a unit time;

[0051] Saturation, which represents the ratio of the actual arriving traffic flow of a lane to the lane’s capacity;

[0052] Queue length refers to the length of the queue of vehicles at the entrance of the intersection, which is divided into the maximum length of the queue of each lane at the entrance of the intersection l max and the average queue length of each lane at the intersection entrance, l;

[0053] Occupancy rate, including spatial occupancy rate Rs and temporal occupancy rate Rt. The spatial occupancy rate indicates the ratio of the total area occupied by vehicles in a specific area of ​​the road to the area of ​​the area. The temporal occupancy rate indicates the ratio of the time occupied by vehicles at a known point on the road to the time interval in a specified time interval.

[0054] Delay time refers to the loss of running time caused by the obstruction of vehicle movement;

[0055] Travel time, which represents the time required for a vehicle to travel a specific road segment;

[0056] Vehicle speed, including the instantaneous speed of a vehicle passing through a certain road section and the average speed of a vehicle passing through a certain area;

[0057] Traffic flow density refers to the number of vehicles per unit length on a lane at a certain point in time;

[0058] The secondary queue rate refers to the ratio of the number of vehicles that did not pass through the intersection in the first signal cycle and entered the secondary queue to the number of all vehicles participating in the queue in the first signal cycle;

[0059] Headway time refers to the time difference when the front ends of two adjacent vehicles pass through the same section.

[0060] Saturation, headway, queue length, delay time and secondary queue rate are selected as characteristic parameters, and an intersection evaluation index model is established through a BP neural network. Traffic flow, travel time, occupancy, traffic flow density and vehicle speed are selected as characteristic parameters, and a section evaluation index model is established through a SVM classifier. The intersection evaluation index model and the section evaluation index model are combined to establish the road network comprehensive evaluation index model.

[0061] In one embodiment, when establishing an intersection evaluation index model through a BP neural network, because there are five characteristic parameters, namely saturation, headway, queue length, delay time and secondary queue rate, as evaluation indicators, a 5-layer BP neural network is used, that is, the input layer has 3 neurons and the output index is 1, that is, the output neuron is one, and the output value range is 0-1. The value indicates the quality of the traffic condition at the intersection. The larger the value, the worse the traffic condition at the intersection and the more congested the intersection. 0-0.25 can be defined as unobstructed, 0.25-0.5 as light congestion, 0.5-0.75 as moderate congestion, and 0.75-1 as heavy congestion. The collected samples are selected to train the BP neural network, and finally the trained model is used to evaluate the road network.

[0062] In one embodiment, when a road section evaluation index model is established by an SVM classifier, five characteristic parameters, namely traffic flow, travel time, occupancy rate, traffic flow density and vehicle speed, are selected as evaluation indicators. The collected data are divided into a training set and a test set according to a certain ratio. The training set sample data and the test set sample data are marked by experts according to the corresponding road section status. In order to correspond to the road conditions at the intersection, the roads are divided into 1-4 according to the operating status, 1 is defined as unobstructed, 2 is defined as light congestion, 3 is defined as moderate congestion, and 4 is defined as heavy congestion. The training model is determined, and the training model is used to train samples to obtain the SVM model. The SVM model is evaluated on the test set samples, and the results are compared with the expert markings to determine whether the accuracy meets the requirements. If the requirements are not met, the training model parameters are returned to adjust until the accuracy of the comparative analysis between the SVM model evaluation results and the expert markings meets the requirements.

[0063] The road network comprehensive evaluation index model is established by combining the intersection evaluation index model and the road section evaluation index model. For example, if a road network includes multiple road sections and intersections, the average score of the intersections in the road network can be calculated through the intersection evaluation index model, and the average score of the road sections in the road network can be calculated through the road section evaluation index model, and then an overall evaluation of the road network can be made. The status of a single road section and intersection can be determined by the model, and the status of the entire road network can be determined by the average, which is convenient for evaluators to compare the status of single road sections and intersections as well as the status between road networks.

[0064] In one embodiment, in order to comprehensively evaluate the road network conditions, a calculation formula may be introduced:

[0065]

[0066] Where: PI(i) is the operating status value of intersection i;

[0067] d is the average delay of vehicles at the intersection, in seconds, and dmax is the maximum parking delay, in seconds;

[0068] Ls is the queue length of the key entrance and exit lanes, in meters, and L is the length of the road section, in meters;

[0069] G is the saturation of the intersection road;

[0070] T is the headway time;

[0071] P is the secondary queuing rate;

[0072] η is a model parameter, and its value range is 0<η<1, and η 1 , η 2 , η 3 , η 4 and η 5 The sum is 1.

[0073] Introduce the calculation formula:

[0074]

[0075] Where: PI(i) is the operating status value of section i;

[0076] V m is the maximum speed limit of road section i, V i is the average travel speed on road section i;

[0077] R m is the maximum occupancy rate of road section i, R i is the average occupancy rate of road section i;

[0078] Q m is the maximum flow rate of road section i, Q i is the average flow rate of road section i;

[0079] t is the travel time of road segment i;

[0080] S m is the maximum traffic flow density of road section i, S i is the average traffic flow density of road section i.

[0081] Then, according to the importance of the road section in the traffic network, an importance weight is attached to each road section. The heavier the traffic task undertaken by the road section in the road network, the greater the corresponding weight. The road sections can be assigned weights of 0.5, 0.3 and 0.2 respectively according to the main road, secondary road and branch road. According to the importance of the road section and the communication capacity and length of the road section, a comprehensive importance weight model of the road section is established:

[0082]

[0083] Where: w(L i ) is the comprehensive weight of the road section;

[0084] w c (L i ) is the importance weight of the road section, which can be expressed as C is the road section capacity, l is the road section length;

[0085] w r (L i ) is the road section weight;

[0086] n is the number of road segments in the road network;

[0087] The weight of the intersection depends on the road segment, so the intersection importance weight model can be established:

[0088]

[0089] Where: w(I i ) is the comprehensive weight of the intersection;

[0090] n is the number of intersections in the road network;

[0091] w r (I i ) is the intersection grade weight model, and the formula is expressed as In this formula, n is the number of road sections connected by the intersection, w r (L i ) is the weight of the road segment connected by the intersection;

[0092] w c (I i ) is the intersection importance weight model, and the formula is expressed as n is the number of road sections connected by the intersection, w c (L i ) is the road segment grade weight.

[0093] Combining the two models, the road network comprehensive evaluation model can be obtained as follows:

[0094]

[0095] Among them, PI is a comprehensive indicator of the road network operation status; β is a model parameter, and its value is [0,1]. The larger the PI value, the worse the road network operation status, and vice versa, the smoother the road network traffic operation. In specific applications, B is adjusted according to the influence of the intersection on the road network traffic operation status.

[0096] Through this model, the road network can be comprehensively evaluated, which is convenient for the evaluation file to comprehensively evaluate each road network, and then determine which road network should be prioritized based on the evaluation results.

[0097] The importance of roads, service demand and socio-economic impact are introduced as characteristic parameters, and the evaluation results of roads by the road network comprehensive evaluation index model are combined to determine the implementation priority ranking table of roads.

[0098] The importance of a road can be judged by factors such as its criticality in the road network, whether it is a major traffic artery or urban passage, and whether it is connected to important areas.

[0099] Service demand can take into account factors such as population density, commercial activities, and public service needs around the road. These factors directly affect the service function of the road, especially in high-demand areas, where the improvement of transportation facilities is particularly important. Based on the evaluation of the number of residents, business activities, and public facility density in the surrounding area, roads in densely populated areas and areas with frequent commercial activities score higher.

[0100] Finally, consider the economic value of the road, its impact on social activities, and the improvement in social benefits after the road is improved, including the potential impact of road improvements on property values ​​in surrounding areas, commercial development, and improved traffic mobility.

[0101] The road network is evaluated again using the above three indicators as priority indicators. The relevant evaluation parameters can be adjusted according to the actual situation. Combined with the aforementioned road network comprehensive evaluation indicator model, a comprehensive ranking is obtained to obtain a comprehensive road implementation priority ranking table, which will help relevant personnel understand which road networks are important and urgently need to be widened and repaired.

[0102] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0103] In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other.

Claims

1. A road network evaluation and road implementation priority ranking method based on multi-source data fusion, characterized in that: include: Establish a road network multi-source data collection system to collect multi-source data of roads in the road network; Preprocess the collected multi-source data, remove noise data, and perform standardization to obtain standardized multi-source data; Integrate various standardized multi-source data to form a comprehensive evaluation index model for the road network; Using the road network comprehensive evaluation index model to evaluate each road, and obtaining an evaluation result for each road; Using a priority sorting algorithm to sort the evaluation results of each road, and determine the implementation priority of the road; The evaluation and ranking results are displayed through a visualization platform for reference by decision makers.

2. The method for road network evaluation and road implementation priority sorting based on multi-source data fusion according to claim 1 is characterized by: The multi-source data collection method includes radar frequency modulation microwave detection, video detection and vehicle-mounted navigation and positioning system detection.

3. The method for road network evaluation and road implementation priority sorting based on multi-source data fusion according to claim 1 is characterized by: The pre-processing comprises: Use threshold method or confidence distance detection method to identify abnormal data; Moving average method, time series method and correlation analysis method were used to fill missing data; The deletion method is used to deal with invalid data that has no use value.

4. The method for road network evaluation and road implementation priority sorting based on multi-source data fusion according to claim 1 is characterized in that: The standardized multi-source data includes: Traffic flow refers to the number of motor vehicles passing through a certain section of road in a unit time; Saturation, which represents the ratio of the actual arriving traffic flow of a lane to the lane’s capacity; Queue length refers to the length of the queue of vehicles at the entrance of the intersection, which is divided into the maximum length of the queue of each lane at the entrance of the intersection l max and the average queue length of each lane at the intersection entrance, l; Occupancy rate, including spatial occupancy rate Rs and temporal occupancy rate Rt. The spatial occupancy rate indicates the ratio of the total area occupied by vehicles in a specific area of ​​the road to the area of ​​the area. The temporal occupancy rate indicates the ratio of the time occupied by vehicles at a known point on the road to the time interval in a specified time interval. Delay time refers to the loss of running time caused by the obstruction of vehicle movement; Travel time, which represents the time required for a vehicle to travel a specific road segment; Vehicle speed, including the instantaneous speed of a vehicle passing through a certain road section and the average speed of a vehicle passing through a certain area; Traffic flow density refers to the number of vehicles per unit length on a lane at a certain point in time; The secondary queue rate refers to the ratio of the number of vehicles that did not pass through the intersection in the first signal cycle and entered the secondary queue to the number of all vehicles participating in the queue in the first signal cycle; Headway time refers to the time difference when the front ends of two adjacent vehicles pass through the same section.

5. The method for road network evaluation and road implementation priority sorting based on multi-source data fusion according to claim 4 is characterized by: Saturation, headway, queue length, delay time and secondary queue rate are selected as characteristic parameters, and an intersection evaluation index model is established through a BP neural network. Traffic flow, travel time, occupancy, traffic flow density and vehicle speed are selected as characteristic parameters, and a section evaluation index model is established through a SVM classifier. The intersection evaluation index model and the section evaluation index model are combined to establish the road network comprehensive evaluation index model.

6. The method for road network evaluation and road implementation priority sorting based on multi-source data fusion according to claim 1 is characterized by: The importance of roads, service demand and socio-economic impact are introduced as characteristic parameters, and the evaluation results of roads by the road network comprehensive evaluation index model are combined to determine the implementation priority ranking table of roads.

Images