CN110796901A - An air traffic situation risk hot spot identification method - Google Patents

Abstract

The invention discloses a method for identifying risk hotspots of air traffic situation. The method has four steps: firstly, the comprehensive track data of aircraft is collected, and the aircraft information at each moment in the air traffic situation is obtained; secondly, based on the positional relationship between the aircrafts, The air traffic situation network is established; then the community structure in the network model is searched to form the initial hotspot sub-region; finally, the initial hotspot sub-region is verified and synthesized to form the air traffic situation risk hotspot. By adopting the method, the aircrafts related to the risk hotspot areas in the air traffic situation can be automatically identified, and the funds are occupied less, and the evaluation method is simple, fast and easy to use.

Description

An air traffic situation risk hot spot identification method

technical field

The invention relates to the field of air traffic management, in particular to an air traffic situation risk hot spot identification method based on the idea of network community structure.

Background technique

In order to reduce the total amount of working memory in the actual air traffic control work, the controller often divides the aircraft into different risk levels according to the space distance between the aircraft and other aircraft. Aircraft that are closer to each other and more prone to collisions are considered more risky aircraft and are divided into different groups based on spatial location. This particular group of aircraft reflects some localized, high-risk areas in the airspace, also known as risk hotspots. The more risk hotspots, the larger the scale and the longer the existence time, the higher the corresponding air traffic situation risk. At present, the controller mainly observes the radar screen with the naked eye, and manually identifies the risk hotspot by judging the distance between the aircraft on the radar screen. This manual identification method is inefficient, and it is not only easy to identify wrong hot spots, but also increases the workload of the controller, which ultimately affects the level of air traffic safety. Therefore, in the operation of air traffic management, it is very necessary to realize automatic identification of risk hotspots through automated systems.

SUMMARY OF THE INVENTION

Aiming at the current lack of an automatic identification method for air traffic situation risk hotspots, the present invention proposes a method for identifying risk hotspots in air traffic situations, which realizes automatic identification of risk hotspots in air traffic situations.

The method consists of four steps: first, collect the comprehensive track data of the aircraft, and obtain the aircraft information at each moment in the air traffic situation; secondly, based on the positional relationship between the aircraft, establish the air traffic situation network; then search for the community structure in the network model, The initial hotspot sub-area is formed; finally, the initial hotspot sub-area is verified to comprehensively form the air traffic situation risk hotspot.

The technical scheme adopted by the present invention is: a method for identifying air traffic situation risk hotspots, the method is assisted by a computer system, and the computer system is mainly composed of a client/server, and is characterized in that the steps are as follows:

Step 1. Lead and process the aircraft track data: According to the original data transmission frequency, receive the aircraft comprehensive track data in real time, including the aircraft's longitude, latitude, altitude, speed, and heading information at each moment;

Step 2. Establish an air traffic situation network model according to the results of Step 1: first, set the horizontal distance threshold as S1 and the vertical distance threshold as S2. At each moment, take the aircraft as a node, and the spatial position relationship between the aircraft is represented by edges , build a network model; set the current moment as t , the total number of aircraft in the air traffic situation at this moment is n , calculate the spatial distance between all aircraft, and generate n × n aircraft horizontal distance matrix D1, vertical distance matrix D2 and the air The adjacency matrix A of the network corresponding to the traffic situation, if the horizontal distance between two aircraft is less than or equal to S1 and the vertical distance is less than S2, that is, there is a close relationship between aircraft i and aircraft j , then a _i,j = 1, otherwise a _{i, j} = 0, taking a _{i, j} as the elements of the adjacency matrix A, the air traffic situation network model at time t can be established;

Step 3. Search the community structure in the network model at each moment to form the initial risk hotspot sub-region: First generate a 1× n access tag array Flag, the initial value of each element in the tag array is 0, and the subsequent process can be subdivided into Follow the steps below:

Step 3.1 Search for a vertex V _i in the adjacency matrix A that satisfies Flag( i )=0. If the search fails, that is, all aircraft are visited, go to step 4, otherwise set Flag( i )=1, and take the node V _i Establish an initial risk hotspot for the starting point search, labeled C ;

Step 3.2 If there is a _i,j = 0 for all vertices V _j (0<j, j≤n, j≠i), that is, the aircraft node has no adjacent aircraft, then go back to step 3.1 to reselect a new node, otherwise continue ;

Step 3.3 Create a new risk hotspot sub-region, marked as c , with initial value c ={ V _i };

Step 3.4 Search all unvisited adjacent aircraft sets V ^' = { V ^' _i ,..., V ^' _j } of node V _i , update the risk hotspot sub-region c = c ∪ V ^' , and set the visit mark of the corresponding node as 1;

Step 3.5 Search all the adjacent aircraft sets V ^{' '} = { V ^'' _i , ..., V ^'' _j } in V ', update the risk hotspot sub-region c = c ∪ V ^'' , set the corresponding The access mark of the node is 1, and the recursive search is performed to finally generate the kth risk hotspot sub-region _ck = c , and return to step 3.1;

Step 4 Verify the initial hotspot sub-regions to form air traffic situation hotspots: first delete the initial risk hotspot sub-regions with the number of nodes less than 3, and further integrate these initial risk hotspot sub-regions to construct the risk in the air traffic situation at time t Hot spot C ( t ), i.e. C ( t ) = { c ( t ) ₁ , c ( t ) ₂ ,..., c ( t ) _i ,..., c ( t ) _k } , where c ( t ) ) _i is the ith risk hotspot sub-region in C ( t ).

The beneficial effects of the invention are: adopting the air traffic situation risk hot spot identification method based on the network community structure idea, can automatically and quickly identify the aircraft related to the risk hot spot area in the air traffic situation, and the risk hot spot identification time in a single sector is less than 2 Second, this method is not affected by human factors, and can effectively reduce the workload caused by the controller's manual judgment of these risk hotspots, does not increase the controller's workload, and occupies less funds, and the evaluation method is simple, fast and easy. The evaluation results are easy to understand, thereby improving the level of air transport safety.

Description of drawings

Fig. 1 is the basic step process flow of the present invention;

FIG. 2 is a schematic diagram of risk hot spots in the air traffic situation of the present invention.

Detailed ways

The present invention will be further described below with reference to the accompanying drawings and embodiments.

An air traffic situation risk hot spot identification method includes the following specific steps, as shown in Figure 1:

Step 1. Lead and process aircraft track data: According to the transmission frequency of the original data, receive the aircraft comprehensive track data in real time, including the longitude, latitude, altitude, speed and heading information of all aircraft in the airspace at each moment;

Step 2. Establish an air traffic situation network model according to the results of Step 1: first, set the horizontal distance threshold as S1 and the vertical distance threshold as S2. At each moment, taking the aircraft as a node, the spatial positional relationship between the aircraft is represented by an edge, and a network model is constructed. Let the current time be t , the total number of aircraft in the air traffic situation at this time is n , calculate the spatial distance between all aircraft, and generate n × n aircraft horizontal distance matrix D1, vertical distance matrix D2 and the corresponding air traffic situation at this moment. The adjacency matrix A of the network. If the horizontal distance between the two aircraft is less than or equal to S1 and the vertical distance is less than S2, that is, there is a close relationship between aircraft i and aircraft j , then a _i,j = 1, otherwise a _i,j = 0. Further, taking a _i,j as the elements of the adjacency matrix A, the air traffic situation network model at time t can be established. Example: If the horizontal distance threshold S1=20 kilometers, the vertical distance threshold S2=600 meters, the horizontal distance between aircraft i and aircraft j is 15 kilometers and the vertical distance is 500 meters, then the distance between aircraft i and aircraft j is 500 meters. There is a proximity relationship between aircraft i _{, j} = 1; if the horizontal distance between aircraft i and aircraft j is 15 kilometers and the vertical distance is 1000 meters, there is no proximity relationship between aircraft i and aircraft j , that is, a _i,j = 0. In practical application, the interval standard issued by the Civil Aviation Administration can be appropriately enlarged and set.

Step 3. Search the community structure in the network model at each moment to form the initial risk hotspot sub-region: First generate a 1× n access tag array Flag, the initial value of each element in the tag array is 0, and the subsequent process can be subdivided into Follow the steps below:

Step 3.1 Search for a vertex V _i in the adjacency matrix A to satisfy Flag( i )=0. If the search fails, ie all aircraft are visited, go to step 4. Otherwise, set Flag( i )= ₁ , and take the node Vi as the starting point to search and establish the initial risk hot spot, marked as C .

Step 3.2 If there is a _i,j = 0 for all vertices V _j (0<j, j≤n, j≠i), that is, the aircraft node has no adjacent aircraft, then go back to step 3.1 to reselect a new node, otherwise continue .

Step 3.3 Create a new risk hotspot sub-region, labeled c , with initial value c = { V _i }.

Step 3.4 Search all unvisited adjacent aircraft sets V ^' = { V ^' _i ,..., V ^' _j } of node V _i , update the risk hotspot sub-region c = c ∪ V ^' , and set the visit mark of the corresponding node as 1.

Step 3.5 Search all the adjacent aircraft sets V ^{' '} = { V ^'' _i , ..., V ^'' _j } in V ', update the risk hotspot sub-region c = c ∪ V ^'' , set the corresponding The access of the node is marked as 1, recursively search, and finally generate the kth risk hotspot sub-region _ck = c , and return to step 3.1 .

Step 4 Verify the initial hotspot sub-regions to form air traffic situation hotspots: first delete the initial risk hotspot sub-regions with the number of nodes less than 3; further integrate these initial risk hotspot sub-regions to construct the risk in the air traffic situation at time t Hot spot C ( t ), i.e. C ( t ) = { c ( t ) ₁ , c ( t ) ₂ ,..., c ( t ) _i ,..., c ( t ) _k } , where c ( t ) ) _i is the ith risk hotspot sub-region in C ( t ).

Figure 2 shows a schematic diagram of the automatic identification results of air traffic situation risk hotspots at a certain time. There are 10 aircraft in this air traffic situation, and there are 10 nodes in the mapped network model. Set the horizontal distance threshold S1 = 20 kilometers, the vertical distance threshold S2 = 600 meters, calculate the horizontal distance and vertical distance between nodes, and compare with S1 and S2. The risk hotspots in this situation include two hotspot sub-areas, denoted as risk hotspot sub-area 1 and risk hotspot sub-area 2. Risk hotspot sub-area 1 includes three nodes N3, N5, and N6, corresponding to three aircrafts P3, P5, and P6 respectively. The horizontal distances between N3 and N5, N3 and N6, and N5 and N6 are The distances are 300, 300, 0, respectively. The risk hotspot sub-area 2 includes three nodes N8, N9, and N10, corresponding to the three aircrafts P8, P9, and P10 respectively. The horizontal distances between N8 and N9, N8 and N10, and N9 and N10 are The distances are 400, 100, 500 respectively. The other four aircraft N1, N2, N4, and N7 in this situation are not risk hotspots.

Claims (1)

1. A method for identifying air traffic situation risk hot spots is realized by assistance of a computer system, wherein the computer system mainly comprises a client/server, and is characterized by comprising the following steps:

step 1, leading and processing aircraft track data: receiving the comprehensive track data of the aircraft in real time according to the sending frequency of the original data, wherein the comprehensive track data comprises longitude, latitude, altitude, speed and course information of the aircraft at each moment;

step 2, establishing an air traffic situation network model according to the result of the step 1: firstly, setting a horizontal distance threshold value as S1 and a vertical distance threshold value as S2, and constructing a network model by taking the aircrafts as nodes and representing the spatial position relation among the aircrafts by edges at each moment; let the current time betThe total number of the aircrafts in the air traffic situation at the moment isnCalculating the space distance between all the aircrafts to generaten×nIf the horizontal distance between two aircrafts is less than or equal to S1 and the vertical distance is less than S2, the aircrafts obtain the horizontal distance matrix D1, the vertical distance matrix D2 and the adjacent matrix A of the network corresponding to the air traffic situation at the momentiAircraft and aircraftjThere is a proximity relationship between them, thena _i,j =1, otherwisea _i,j =0, toa _i,j For elements of the adjacency matrix A, i.e. thetA temporal air traffic situation network model;

step 3, searching a community structure in the network model at each moment to form an initial risk hotspot sub-region: first generate 1nThe initial value of each element in the tag array is 0, and the following process can be subdivided into the following steps:

step 3.1 search for a vertex in the adjacency matrix AV _i Satisfies Flag (f)i) =0, if the search fails, i.e. all aircraft are visited, go to step 4, otherwise set Flag: (1)i) =1, in nodesV _i Establishing initial risk for an origin searchHot spots, marked asC；

Step 3.2 if for all verticesV _j (0<j, j is less than or equal to n, j is not equal to i) are alla _i,j =0, if the aircraft node is not adjacent to the aircraft, returning to the step 3.1 to reselect a new node, otherwise, continuing;

step 3.3 create a new risk hotspot sub-area, marked ascInitial value ofc={V _i }；

Step 3.4 search for nodesV _i Set of all non-visited contiguous aircraftV ^’={V ^’ _i,...,V ^’ _jAnd updating risk hotspot sub-areasc=c∪V ^’Setting an access mark of a corresponding node as 1;

step 3.5 searchV ^’Set of contiguous aircraft of all unvisited aircraftV ^’’= {V ^’’ _i , ...,V ^’’ _j And updating risk hotspot sub-areasc=c∪V ^’’Setting the access flag of the corresponding node to 1, performing recursive search, and finally generating the first nodekIndividual risk hot spot sub-areac _k =cReturning to the step 3.1;

step 4, verifying the initial hotspot sub-area to form an air traffic situation hotspot: firstly, deleting initial risk hot spot sub-regions with the number of nodes less than 3, and further integrating the initial risk hot spot sub-regions to constructtRisk hot spot in air traffic situation at momentC(t) I.e. byC(t) = {c(t)₁,c(t)₂,...,c(t) _i ,...,c(t) _k Therein ofc(t) _i Is thatC(t) To middleiIndividual risk hot spot sub-regions.

Images