A method for aviation network risk assessment
By obtaining airport risk time series and evaluating airport saturation, a geographical cluster of risk airports is formed, and the problem of insufficient assessment of potential safety risks of flight delays and airport saturation in the existing technology is solved, and dynamic risk assessment and prevention of aviation networks is realized The early implementation of measures has improved the reliability of the assessment.
Patent Information
- Application Number
- CN202010591348.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-06-24
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2040-06-24
AI Technical Summary
The existing aviation network risk assessment methods mainly focus on airline maintenance policies, human errors and aircraft operations. They lack a systematic assessment of the potential safety risks brought about by flight delays and airport saturation, and the evaluation factors are subjective and difficult to Prevent major flight accidents in advance.
By obtaining the risk time series of the airport, dividing the risk and non-risk airport sets, forming a geographical cluster of risk airports, and obtaining the airspace risk quantification value based on the airspace range, conducting airspace risk ratings, selecting airport saturation and flight delays as evaluation factors, and in advance Adjust strategies to reduce the risk of major accidents.
It has achieved control over the overall operating situation of the airspace, prevented major flight accidents in advance, reduced high-risk operations at the tactical level, and improved the reliability and dynamic nature of aviation network risk assessment.
Smart Images

Figure CN112036679B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to an aviation network risk assessment method, belonging to the technical field of civil aviation risk assessment. Background Art
[0002] With the development of my country's air transport industry, the increasing demand for flights will bring tremendous pressure to the aviation network, which is already unable to cope with the current demand. Due to the large increase in the number of aircraft in limited airspace, the delays and congestion of the aviation network will be further aggravated. Delays and congestion of the aviation network are important potential risks in civil aviation safety management. If there are potential safety risks in the aviation network and it is in a high-risk state, it will pose a major threat to civil aviation safety, which may lead to fluctuations in the level of civil aviation safety and even worsen the civil aviation situation.
[0003] In terms of aviation network risk assessment, existing research on aviation network risk assessment usually focuses on airline maintenance policies, human errors, aircraft operations, etc. It mainly starts from accidents and safety indicators to assess the main causes and probabilities of aircraft flight accidents, or is committed to investigating accidents caused by human factors and establishing causal models. Such research emphasizes analyzing the precursors of past flight accidents and exploring corrective measures to prevent future errors. There are major drawbacks in risk assessment based on flight accident precursors. Once a flight accident occurs, it is the highest level of risk. For such risk assessment, the adjustment strategy can only be targeted at the tactical level, which is difficult to operate and has a short response time.
[0004] In addition, the aviation network is a large and complex system with a low probability of flight accidents and many influencing factors. The selection of existing influencing factors is based on low-probability events and often relies on expert opinions, which is somewhat subjective. Summary of the invention
[0005] The technical problem to be solved by the present invention is to provide an aviation network risk assessment method.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] An aviation network risk assessment method comprises the following steps:
[0008] Obtain the corresponding risk time series of each airport in the aviation network, and divide the risk airport set and non-risk airport set corresponding to each time interval according to a given airport risk quantification coefficient threshold; wherein the corresponding risk time series of each airport includes the corresponding risk quantification coefficient of each airport in all time intervals within a specific time period;
[0009] According to the risk airport set corresponding to a certain time interval and the route connection relationship between airports, the risk airport geographical cluster of the aviation network in this time interval is obtained;
[0010] According to the airspace range, the geographical clusters of risky airports within a certain time interval are cut to obtain the risky airport clusters in each airspace within the time interval;
[0011] Based on the risk airport geographical clusters and the risk quantification coefficients of the airports in each airspace within a certain time interval, the airspace risk quantification value of each airspace within the time interval is obtained, and then an airspace risk rating is performed.
[0012] Preferably, obtaining the corresponding risk time series of each airport in the aviation network, and dividing the risk airport set and the non-risk airport set corresponding to each time interval according to the given airport risk quantification coefficient threshold, comprises the following steps:
[0013] Obtaining the average take-off delay time and airport saturation of all time intervals of each airport within a specific time period;
[0014] After normalizing the average takeoff delay time and the airport saturation of all airports in a certain time interval, clustering is performed to obtain m takeoff delay intervals and saturation intervals and m-1 takeoff delay time cutoff values and airport saturation cutoff values, where m is a positive integer;
[0015] The average takeoff delay time and the airport saturation of each airport in a certain time interval are matched to corresponding takeoff delay intervals and saturation intervals, and are respectively mapped into corresponding delay weight vectors and saturation weight vectors through corresponding mapping functions based on Gaussian kernel functions;
[0016] According to the delay weight vector and the saturation weight vector of each airport in all time intervals, a risk quantification coefficient of each airport in all time intervals is obtained to form a corresponding risk time series of each airport;
[0017] According to the given airport risk quantification coefficient threshold, the risk airport set and non-risk airport set corresponding to each time interval are divided.
[0018] Preferably, the average take-off delay time of all time intervals at each airport in a specific time period is obtained according to the following formula:
[0019]
[0020] Among them, d i (t) is the average takeoff delay time of airport i in the time interval from t to t+1, Di(t) is the total takeoff delay time of airport i in the time interval from t to t+1; Ci(t) is the number of cancelled flights at airport i in the time interval from t to t+1; 3 is the equivalent takeoff delay time of cancelled flights; P i(t) is the total number of scheduled departure flights from airport i in the time interval from t to t+1; t is a positive integer, representing the tth time interval.
[0021] Preferably, the airport saturation of each airport in all time intervals within a specific time period is: the ratio of the inbound flow of airport i to the number of aircraft entering airport i in the time interval from t to t+1.
[0022] Preferably, the delay weight vector of each airport in a certain time interval is expressed as:
[0023]
[0024] Where Wd(i,t) is the delay weight vector of airport i in the time interval from t to t+1, d i (t) is the average takeoff delay time of airport i in the time interval from t to t+1, k is the takeoff delay interval number in the time interval from t to t+1, j is all the takeoff delay intervals before the current takeoff delay interval k; n is the number of takeoff delay intervals in the time interval from t to t+1 obtained by clustering;
[0025]
[0026] Among them, W s (i, t) is the saturation weight vector of airport i in the time period from t to t+1; s i (t) are the saturation values of airport i in the time period from t to t+1, h is the saturation interval number in the time period from t to t+1, v is all saturation intervals before the current saturation interval h; n is the number of saturation intervals in the time period from t to t+1 obtained by clustering.
[0027] Preferably, the corresponding risk quantification coefficient of all time intervals of each airport in a specific time period is the modulus of a two-dimensional plane vector synthesized by the delay weight vector and the saturation weight vector of the airport in the corresponding time interval.
[0028] Preferably, the process of forming the risk airport geographical cluster of the aviation network within a certain time interval includes the following steps:
[0029] A risk airport is randomly selected from the risk airport cluster in a certain time interval as the starting airport of a risk airport geographical cluster, and related airports of the starting airport in the time interval are determined;
[0030] Check whether the related airports of the selected risk airport are risk airports. If so, they are incorporated into the risk airport geographical cluster of the selected time interval. Otherwise, they are defined as isolated airports.
[0031] Check whether the related airports of each risk airport incorporated into the risk airport geographical cluster are risk airports. If so, merge them into the risk airport geographical cluster until all the related airports of all risk airports newly incorporated into the risk airport geographical cluster are quarantine airports, and then obtain a risk airport geographical cluster for the selected time interval;
[0032] The risk airports in the risk airport concentration of the selected time interval that do not belong to the determined risk airport geographical clusters will continue to be the starting airports of the next risk airport geographical cluster, and their related airports will be confirmed and it will be determined whether the related airports are dangerous airports, until all the airports in the risk airport concentration that can form risk airport geographical clusters are incorporated into the corresponding risk airport geographical clusters.
[0033] Preferably, after obtaining the airspace risk quantification value of each airspace in the time interval according to the risk airport geographical cluster in each airspace in the time interval, the process of performing airspace risk rating comprises the following steps:
[0034] According to the risk airport geographical clusters in each airspace within a certain time interval, the airport risk and cascade risk of each airspace within the time interval are obtained;
[0035] Obtaining a quantitative value of airspace risk for each airspace according to the cascade risk and the airport risk of each airspace within a certain time interval;
[0036] Obtain the airspace risk threshold within a certain time interval and divide the airspace risk level;
[0037] According to the airspace risk level in a certain time interval, a risk rating is performed on each airspace in the time interval.
[0038] Preferably, the airport risk of each airspace in a certain time interval is obtained by taking the weighted average of the risk quantification coefficient values of the risk airports in each airspace;
[0039] The cascade risk of each airspace in a certain time interval is obtained according to the following formula:
[0040]
[0041] Among them, R l (w,t) is the cascade risk of a certain space w in the time interval from t to t+1, l j(w,t) ,j=0,…,q w is the number of risk airports in the jth risk airport geographical cluster in airspace w, m w is the total number of airports in airspace w, q w is the number of risk airport geographical clusters within the airspace w.
[0042] Preferably, the quantified value of airspace risk of a certain airspace within a certain time interval is obtained according to the following formula.
[0043]
[0044] Where α = m w / s, is the airport density of airspace w, m w is the total number of airports in airspace w, s is the total number of airports in the entire aviation network, β is the route density of airspace w and u w is the total number of routes in airspace w, R c (w,t) is the airport risk of a certain airspace w in the time interval from t to t+1, R l (w,t) is the cascading risk of the spatial domain w in the time interval from t to t+1.
[0045] Preferably, the process of obtaining the airspace risk threshold within a certain time interval is:
[0046] The airport risk threshold and the airspace cascade risk threshold within a certain time interval are obtained respectively;
[0047] According to the airport risk threshold and the airspace cascade risk threshold in a certain time interval, the airspace risk threshold of the time interval is obtained, and the airspace risk level is divided.
[0048] Preferably, the process of obtaining the airport risk threshold within a certain time interval is:
[0049] Respectively obtaining the average take-off delay time threshold and the airport saturation threshold within a certain time interval;
[0050] The average takeoff delay time threshold and the airport saturation threshold in a certain time interval are matched to the corresponding takeoff delay interval and saturation interval, and are respectively mapped into corresponding delay weight threshold vectors and saturation weight threshold vectors through corresponding mapping functions based on Gaussian kernel functions;
[0051] According to the delay weight threshold vector and the saturation weight threshold vector in a certain time interval, obtaining an airport risk threshold in the time interval;
[0052] When obtaining the airspace cascade risk threshold within a certain time interval, first obtain the airspace cascade risk value set of each airspace in different time intervals within the past preset time, and then perform clustering to obtain the airspace cascade risk threshold for a certain time interval.
[0053] The aviation network risk assessment method provided by the present invention has the following advantages over the prior art:
[0054] 1. Based on the risks of a single airport, studying the more serious airspace risks caused by risk cascades and conducting airspace-oriented risk assessments will help control the overall operating situation of the airspace.
[0055] 2. Potential safety risks such as airport saturation and flight delays are selected as assessment factors. Starting from the factors that cause aviation network risks in the pre-tactical stage, the time points for risk assessment and adjustment measures are advanced to prevent them from developing into major flight accidents such as collisions, thereby reducing high-risk operations at the tactical level.
[0056] 3. Flight delays and airport saturation are dynamic, which makes the aviation network security risk assessment based on delays and congestion dynamic, making the risk level assessment of each airspace in the entire aviation network more reliable. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] Figure 1 A simplified flowchart of the aviation network risk assessment method provided by the present invention;
[0058] Figure 2 A detailed flowchart of the aviation network risk assessment method provided by the present invention;
[0059] Figure 3 In the aviation network risk assessment method provided by the present invention, a geographical cluster evolution diagram of a certain risk airport;
[0060] Figure 4 An airspace risk level classification diagram is provided in the aviation network risk assessment method provided by the present invention. DETAILED DESCRIPTION
[0061] The technical content of the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0062] With the development of the aviation transportation industry, the situation of aviation network congestion and delay caused by the main contradiction will become more severe, and the potential risks brought by itself and its spread will become a major risk threatening the security of the aviation network. However, the existing aviation network risk assessment methods rarely start from the potential safety risks caused by delays and congestion. In addition, for the phenomenon of aviation network delays and congestion, more studies are conducted on delays and congestion and their spillover effects, focusing on a single airport or flight. The spillover phenomenon is not studied from the perspective of airspace, and delays and congestion are rarely used to quantify risks.
[0063] On the other hand, the airports in the aviation network are composed of highly interconnected airport clusters. There are multiple resource connections between airports due to a large number of flights. The potential safety risks caused by delays and congestion at a single airport will spread among multiple related airports, causing large-scale delays and congestion within a certain airspace, and even causing large-scale paralysis of the airport network, becoming a system-level risk, increasing the operating load of the air traffic system, and increasing safety hazards and flight conflicts.
[0064] In order to overcome the above defects, Figure 1 As shown, the present invention provides an aviation network risk assessment method, which mainly includes the following steps:
[0065] Step S1: Obtain the corresponding risk time series of each airport in the aviation network, and divide the risk airport set and non-risk airport set corresponding to each time interval according to the given airport risk quantification coefficient threshold; wherein the corresponding risk time series of each airport includes the corresponding risk quantification coefficient of each airport in all time intervals within a specific time period.
[0066] Step S2: Based on the risk airport set corresponding to a certain time interval and the route connection relationship between airports, the risk airport geographical cluster of the aviation network in the time interval is obtained.
[0067] Step S3: Cut the risk airport geographical clusters within a certain time interval according to the airspace range to obtain the risk airport geographical clusters in each airspace within a certain time interval.
[0068] Step S4: Based on the risk airport geographical clusters and the risk quantification coefficients of the airports in each airspace within a certain time interval, the airspace risk quantification value of each airspace within the time interval is obtained, and then the airspace risk rating is performed.
[0069] In order to enable those skilled in the art to better understand the present invention, the following describes the implementation of the present invention in conjunction with specific embodiments. Figure 2 As shown, an embodiment of the present invention provides an aviation network risk assessment method, which specifically includes the following steps:
[0070] Step 100: Obtain the average take-off delay time and airport saturation of all time intervals of each airport in a specific time period.
[0071] According to the historical flight schedule data of all airports in the aviation network obtained from the Civil Aviation Administration, air traffic control data companies or travel websites, and the following formulas (1) and (2), the average takeoff delay time and airport saturation of all time intervals of each airport in a specific time period are obtained. Among them, all time intervals in a specific time period are set according to actual needs. For example, a certain day can be selected as a specific time period. In a certain day time period, 1 hour is a time interval. Then the specific time period of the day can be divided into 24 time intervals, that is, the average takeoff delay time and airport saturation of all airports in the aviation network are calculated once every 1 hour. It is also possible to select 1 hour as a specific time period. In this specific time period, 5 minutes is a time interval. Then the specific time period can be divided into 12 time intervals, that is, the average takeoff delay time and airport saturation of all airports in the aviation network are calculated once every 5 minutes. Therefore, the specific time period and all time intervals can be set according to actual needs, and they will not be repeated here.
[0072]
[0073] Among them, d i P(t) is the average departure delay time of airport i during the time interval from t to t+1, Di(t) is the total departure delay time of airport i during the time interval from t to t+1; Ci(t) is the number of cancelled flights at airport i during the time interval from t to t+1; 3 is the equivalent departure delay time of cancelled flights, that is, every time a flight is cancelled, the flight is equivalent to a departure delay of 3 hours; P i (t) is the total number of scheduled departure flights from airport i in the time interval from t to t+1; t is a positive integer, representing the tth time interval.
[0074] Specifically, when calculating the average takeoff delay time of airport i in the time interval from t to t+1, it is necessary to retrieve the historical flight schedule data of airport i in the time interval from t to t+1 from the historical flight schedule data of all airports in the aviation network. The historical flight schedule data includes the actual takeoff time and planned takeoff time of all aircraft at airport i, the number of canceled flights and the total number of planned takeoff flights. Among them, according to the actual takeoff time and planned takeoff time of all aircraft at airport i in the time interval from t to t+1, the total takeoff delay time of airport i can be calculated.
[0075]
[0076] Among them, si(t) is the saturation of airport i in the time interval from t to t+1; Qi(t) is the inbound traffic of airport i in the time interval from t to t+1, that is, the number of aircraft entering airport i in the time interval from t to t+1; Ei(t) is the airport capacity of airport i in the time interval from t to t+1, that is, the maximum number of aircraft entering and landing at airport i that can be tolerated in the time interval from t to t+1; t is a positive integer, representing the tth time interval.
[0077] Step S101: After normalizing the average takeoff delay time and airport saturation of all airports in a certain time interval, clustering is performed to obtain m takeoff delay intervals and saturation intervals and m-1 takeoff delay time cutoff values and airport saturation cutoff values, where m is a positive integer.
[0078] Specifically, taking the average takeoff delay time and airport saturation of each airport in the time interval from t to t+1 as an example, four takeoff delay intervals and saturation intervals and three takeoff delay time cutoff values and airport saturation cutoff values are obtained. After data normalization processing is performed on the average takeoff delay time and airport saturation of all airports in the time interval from t to t+1, clustering processing is performed to obtain four takeoff delay intervals ([0,λ d,1 ],[λ d,1 ,λ d,2 ],[λ d,2 ,λ d,3 ],[λ d,3 , max(di(t))]) and the three takeoff delay time cutoff values (λ d,1 ,λ d,2 ,λ d,3 ), 4 saturation intervals ([0,λ s,1 ,],[λ s,1 ,,λ s,2 ],[λ s,2 ,λ s,3 ],[λ s,3 ,max(si(t))]) and three airport saturation cutoff values (λs,1,λs,2,λs,3). Among them, max(di(t)) is the largest value among the average takeoff delay time of all airports in the time interval from t to t+1, and max(si(t)) is the largest value among the airport saturation of all airports in the time interval from t to t+1. Among them, the number of takeoff delay intervals and saturation intervals, takeoff delay time cutoff values and airport saturation cutoff values in a certain time interval depends on actual needs.
[0079] Step S102: The average takeoff delay time and airport saturation of each airport in a certain time interval are mapped to corresponding takeoff delay intervals and saturation intervals, and are respectively mapped into corresponding delay weight vectors and saturation weight vectors through corresponding mapping functions based on Gaussian kernel functions.
[0080] The delay weight vector of each airport in a certain time interval is expressed as:
[0081]
[0082] Where Wd(i,t) is the delay weight vector of airport i in the time interval from t to t+1, d i (t) is the average takeoff delay time of airport i in the time interval from t to t+1, k is the takeoff delay interval number in the time interval from t to t+1, j is all the takeoff delay intervals before the current takeoff delay interval k; n is the number of takeoff delay intervals in the time interval from t to t+1 obtained by clustering. For example, when n=4, it means that there are 4 takeoff delay intervals and 3 takeoff delay time boundaries in the time interval from t to t+1. At this time, λ d,k is the takeoff delay time boundary value in the time interval from t to t+1, respectively λ d,1 ,λ d,2 ,λ d,3 ;
[0083] The saturation weight vector of each airport in a certain time interval is expressed as:
[0084]
[0085] Among them, W s (i, t) is the saturation weight vector of airport i in the time period from t to t+1; s i (t) are the saturation values of airport i in the time interval from t to t+1, h is the saturation interval number from t to t+1, v is all saturation error intervals before the current saturation interval h; n is the number of saturation intervals in the time interval from t to t+1 obtained by clustering. For example, when n=4, it means that there are 4 saturation intervals and 3 airport saturation boundary values in the time interval from t to t+1. At this time, λ s,k is the airport saturation threshold value in the time interval from t to t+1, respectively λ s,1 ,λ s,2 λ s,3 .
[0086] Step S103: According to the delay weight vector and saturation weight vector of each airport in all time intervals, the risk quantification coefficient of each airport in all time intervals is obtained to form the corresponding risk time series of each airport.
[0087] According to the delay weight vector and saturation weight vector of each airport in a certain time interval obtained in step S102, the risk quantification coefficient of each airport in a certain time interval is obtained using the following formula.
[0088] R i (t)=||W d (i,t)+W s (i,t)|| (5)
[0089] Among them, R i (t) is the risk quantification coefficient of airport i in the time interval from t to t+1, W d (i, t) is the delay weight vector of airport i in the time interval from t to t+1, W s (i, t) is the saturation weight vector of airport i in the time period from t to t+1; that is, the risk quantification coefficient of airport i in the time period from t to t+1 is the modulus of the two-dimensional plane vector synthesized by the delay weight vector and the saturation weight vector of the airport in the time period from t to t+1.
[0090] The risk quantification coefficients are calculated for all time intervals of all airports in the aviation network within a preset time period to form the corresponding risk time series for each airport.
[0091] Step S104: Divide each time interval into a risk airport set and a non-risk airport set according to a given airport risk quantification coefficient threshold.
[0092] If the risk quantification coefficient of an airport in the time interval from t to t+1 is greater than or equal to the given airport risk quantification coefficient threshold R0, that is, R i (t)≥R0, then the airport is judged as a risk airport in the time interval from t to t+1, and the airport is added to the risk airport set in the time interval from t to t+1. On the contrary, if the risk quantification coefficient value of a certain airport is lower than the given airport risk quantification coefficient threshold R0, then the airport is judged as a non-risk airport. Repeat step S104 to complete the risk judgment of all airports from t to t+1, and obtain the risk airport set and non-risk airport set in the time interval from t to t+1. Using the judgment method of the risk airport set and non-risk airport set in the time interval from t to t+1, the risk airport set and non-risk airport set corresponding to each time interval can be obtained.
[0093] Step S105: randomly selecting a risk airport from the risk airport cluster in a certain time interval as the starting airport of the risk airport geographical cluster, and determining the related airports of the starting airport in the time interval.
[0094] Related airports are defined using the connection relationship between routes between airports in the airspace. Airports that are connected to a certain airport by routes are called related airports of this airport.
[0095] like Figure 3 As shown in the figure, taking the formation process of a certain risk airport geographical cluster in the time interval from t to t+1 as an example, a risk airport a is randomly selected from the risk airport set in the time interval from t to t+1 obtained in step S50 as the starting airport of the risk airport geographical cluster, and the related airports of risk airport a are determined to be {a1, a2, …, a i ,…,a h}, total h.
[0096] Step S106: Check whether the related airports of the selected risk airport are risk airports. If so, they are incorporated into the risk airport geographical cluster of the selected time interval. Otherwise, they are defined as isolated airports.
[0097] like Figure 3 As shown, check whether h related airports are risk airports in turn, for example, airport a i As a risk airport, airport a i Merge into the risk airport geographical cluster, otherwise airport a i Defined as a quarantine airport.
[0098] Step S107: Check whether the related airports of each risk airport incorporated into the risk airport geographical cluster are risk airports. If so, incorporate them into the risk airport geographical cluster until the related airports of all risk airports newly incorporated into the risk airport geographical cluster are all isolated airports, and a risk airport geographical cluster is obtained for the selected time interval.
[0099] like Figure 3 As shown in the figure, if there are s risk airports among the h related airports of risk airport a, then these s risk airports are incorporated into the risk airport geographical cluster, and the related airports of the s risk airports incorporated this time are continuously checked to see if they are risk airports. If so, they are continuously incorporated into the risk airport geographical cluster, otherwise they are defined as isolated airports. If there are c risk airports among the related airports of s risk airports, then these c risk airports are incorporated into the risk airport geographical cluster, and the related airports of the c risk airports incorporated this time are continuously checked to see if they are risk airports. If all the related airports of the c risk airports are isolated airports, a risk airport geographical cluster is obtained for the time interval from t to t+1.
[0100] Step S108: The risk airports in the risk airport concentration of the selected time interval that do not belong to the determined risk airport geographical cluster will continue to be the starting airports of the next risk airport geographical cluster, and their related airports will be confirmed and it will be determined whether the related airports are dangerous airports, until all the airports in the risk airport concentration that can form risk airport geographical clusters are incorporated into the corresponding risk airport geographical clusters.
[0101] Since one or more risk airports may appear in the risk airport set in the selected time interval that are not related airports of the selected risk airport and cannot belong to the same risk airport geographical cluster as the selected risk airport; therefore, it is necessary to include the risk airports in the risk airport set in the selected time interval that do not belong to the determined risk airport geographical cluster, and continue to execute step S106 until all the risk airports in the risk airport set are incorporated into the corresponding risk airport geographical cluster, thereby obtaining one or more risk airport geographical clusters in the time interval from t to t+1.
[0102] Step S109: Use the flight information region to define the airspace scope of the aviation network, and geographically segment the risk airport geographical clusters within a certain time interval according to the airspace scope to obtain the risk airport clusters in each airspace.
[0103] "Flight Information Region" refers to the space demarcated for providing flight information services and warning services for air traffic management business. Evaluating the aviation network risk in each flight information region is conducive to the air traffic control department to learn the current risk status of air traffic in the region and the special risk information that needs attention. Therefore, in the present invention, the use of flight information region division standards to divide the entire aviation network into geographical airspaces has important practical significance and wide application value. A total of 11 flight information regions are divided in the country, that is, the entire aviation network is divided into 11 airspace ranges. According to the 11 airspace ranges and the locations of the airports distributed therein, the geographical clusters of risk airports in a certain time interval are geographically cut, and all the geographical clusters of risk airports in the time interval are divided into corresponding airspaces according to their specific geographical locations. That is, the geographical location of the risk airport geographical cluster in the selected time interval belongs to which airspace, and it is divided into the corresponding airspace accordingly, so as to obtain the risk airport cluster in each airspace in a certain time interval.
[0104] Step S110: Based on the risk airport geographical clusters in each airspace within a certain time interval, the airport risk and cascade risk of each airspace within the time interval are obtained.
[0105] The airport risk of each airspace in a certain time interval is obtained according to the following formula.
[0106]
[0107] Among them, R c (w,t) is the airport risk of a certain airspace w in the time interval from t to t+1, R i(w,t) (i=0,…,n w ) is the risk quantification coefficient value of risk airport i in the time interval from t to t+1 in airspace w, which can be obtained from the corresponding risk time series of each airport obtained in step S103. w is the number of risk airports in airspace w. According to the risk airport set corresponding to each time interval obtained in step S104, the airspace to which each airport in the risk airport set belongs is determined, thereby obtaining the number of risk airports in each airspace; E wi is the degree of risk airport i, that is, the number of routes related to the risk airport; E w is the number of relevant routes of all airports in airspace w (including non-risky ones). It is not difficult to find that the airport risk of each airspace in a certain time interval can be obtained by taking the weighted average of the risk quantification coefficient values of risky airports in each airspace.
[0108] Based on the size and number of all risk airport geographical clusters in each airspace in a certain time interval, the cascade risk of each airspace is obtained according to the following formula.
[0109]
[0110] Among them, R l (w,t) is the cascade risk of a certain space w in the time interval from t to t+1, l j(w,t) ,j=0,…,q w is the number of risk airports in the jth risk airport geographical cluster in airspace w, m w is the total number of airports in airspace w, q w is the number of risk airport geographical clusters within the airspace w.
[0111] Step S111: According to the cascade risk and airport risk of each airspace within a certain time interval, a quantitative value of the airspace risk of each airspace is obtained.
[0112] The quantitative value of airspace risk of a certain airspace within a certain time interval is obtained according to the following formula.
[0113]
[0114] Where α = m w / s, is the airport density of airspace w, m w is the total number of airports in airspace w, s is the total number of airports in the entire aviation network, β is the route density of airspace w and u w is the total number of routes in airspace w, R c(w,t) is the airport risk of a certain airspace w in the time interval from t to t+1, R l (w,t) is the cascading risk of the space w in the time interval from t to t+1. Figure 4 As shown, the values of airport risk and cascade risk of airspace w in the time interval from t to t+1 are both taken as 0 as the origin. At this time, the quantified value of airspace risk of airspace w in a certain time interval is 0.
[0115] Step S112: Obtain the airspace risk threshold within a certain time interval and divide the airspace risk levels.
[0116] The process of obtaining the airspace risk threshold within a certain time interval is:
[0117] Step S1120: Obtain the airport risk threshold and the airspace cascade risk threshold within a certain time interval respectively.
[0118] Furthermore, the process of obtaining the airport risk threshold within a certain time interval is as follows:
[0119] Step S11201: respectively obtain an average take-off delay time threshold and an airport saturation threshold within a certain time interval.
[0120] According to the "Airline Flight Delay Warning Management Model and Analysis" (published in "Computer Simulation" 2009 Issue 04) and the flight delay standards stipulated in the "Flight Normal Management Regulations", the average take-off delay time of all airports in a certain time interval in step 100 is graded to obtain three average take-off delay time thresholds. For example, the average take-off delay time thresholds of each airport in the time interval from t to t+1 are 40 minutes, 60 minutes and 120 minutes respectively. According to the three average take-off delay time thresholds, four delay levels can be obtained, and the delay status corresponding to each delay level is respectively expressed as slight delay, medium delay, serious delay and dangerous delay. Specifically, if the average delay time of a certain airport in the time interval from t to t+1 is less than 15 minutes, it is considered that there is no delay at this airport. If the average delay time of a certain airport in the time interval from t to t+1 is between 15 and 40 minutes, it is considered that there is a slight delay at this airport. If the average delay time of an airport in the time interval from t to t+1 is between 40 and 60 minutes, it is considered that the airport has moderate delay. If the average delay time of an airport in the time interval from t to t+1 is between 60 and 120 minutes, it is considered that the airport has severe delay.
[0121] Referring to "Research on Road Saturation Calculation Method" (published in "Traffic Standardization" No. 01, 2007), the airport saturation of all airports in a certain time interval in step 100 is graded to obtain three airport saturation thresholds. For example, the airport saturation thresholds in the time interval from t to t+1 are 6, 0.8 and 1 respectively. According to the three airport saturation thresholds, four airport saturation levels can be obtained. The delay status corresponding to each airport saturation level is respectively expressed as high operation level, good operation level, poor operation level and extremely poor operation level. Specifically, if the airport saturation of an airport in the time interval from t to t+1 is less than 0.6, it is considered that the airport is in a low saturation state. If the airport saturation of an airport in the time interval from t to t+1 is between 0.6 and 0.8, it is considered that the airport is in a low and slightly saturated state. If the airport saturation of an airport in the time interval from t to t+1 is between 0.8 and 1, it is considered that the airport is in a saturated state. If the airport saturation of an airport in the time interval from t to t+1 is greater than 1, it is considered that the airport is in a serious saturation state.
[0122] Step S11202: The average takeoff delay time threshold and the airport saturation threshold within a certain time interval are mapped to the corresponding takeoff delay interval and saturation interval, and are respectively mapped into corresponding delay weight threshold vectors and saturation weight threshold vectors through corresponding mapping functions based on Gaussian kernel functions.
[0123] After the three average takeoff delay time thresholds and the three airport saturation thresholds within a certain time interval obtained in step S11201 are matched to the corresponding takeoff delay interval and saturation interval in step S101, three delay weight threshold vectors and three saturation weight threshold vectors within a certain time interval are obtained according to formulas (3) and (4).
[0124] Step S11203: According to the delay weight threshold vector and the saturation weight threshold vector in a certain time interval, the airport risk threshold in the time interval is obtained.
[0125] According to the three delay weight threshold vectors and the three saturation weight threshold vectors in a certain time interval and formula (5), the three airport risk thresholds Q(i, t) in the time interval are obtained, i = 1, 2, 3.
[0126] Furthermore, the process of obtaining the airspace cascade risk threshold within a certain time interval is as follows:
[0127] Step S11204: After obtaining the airspace cascade risk value sets of each airspace in different time intervals within the past preset time, clustering is performed to obtain the airspace cascade risk threshold of a certain time interval.
[0128] If it is necessary to evaluate the airspace risk in a certain time interval and obtain the airspace cascade risk threshold of the time interval, the airport and airspace data of the preset time (such as 12 months) in the past month of the time interval can be selected. The airport and airspace data include the total number of airports and risk airport geographical clusters in each airspace and the number of risk airports in each risk airport geographical cluster. As the development trend of civil aviation is constantly updated, the airspace cascade risk threshold is also based on dynamically updated big data. According to formula (7), after obtaining the airspace cascade risk value set of each airspace in different time intervals within the preset time in the past, clustering is performed to obtain three airspace cascade risk thresholds P(j,t) for a certain time interval, j=1,2,3.
[0129] Step S1121: According to the airport risk threshold and the airspace cascade risk threshold within a certain time interval, the airspace risk threshold of a certain time interval is obtained, and the airspace risk level is divided.
[0130] The modulus of the three two-dimensional vectors composed of the airport risk threshold Q(i, t) and the airspace cascade risk threshold P(j, t), j = 1, 2, 3 in a certain time interval is obtained, and the three final airspace risk thresholds Rz i(t), i = 1, 2, 3 are obtained, and four airspace risk intervals and risk levels are divided. Therefore, the airport risk threshold and airspace cascade risk threshold in each time interval can be obtained, thereby obtaining the airspace risk threshold in each time interval and dividing the airspace risk level in each time interval.
[0131] Step S113: According to the airspace risk level of a certain time interval, a risk rating is performed on each airspace in the time interval.
[0132] According to the four airspace risk intervals and risk levels of a certain time interval obtained in step S1121, and the airspace risk quantification value of each airspace in a certain time interval obtained in step S111, a risk rating is performed on each airspace in the time interval. Therefore, according to the airspace risk level of each time interval and the airspace risk quantification value of all airspaces in each time interval, a risk rating can be performed on each airspace in each time interval, forming a risk evolution state of all airspaces over time.
[0133] Taking the risk level assessment of the airspace w in the time interval from t to t+1 as an example, the airspace risk quantification value R of the airspace w in the time interval from t to t+1 obtained in step S111 is z (w, t), as well as the four airspace risk intervals and risk levels divided by the airspace risk threshold Rzi(t), i = 1, 2, 3 in the time interval from t to t+1, to evaluate the risk level of airspace w.
[0134] If 0 <R z (w,t) <Rz1 (t), this airspace w is at low risk.
[0135] If R z1 (t) <R z (w,t) <R z2 (t), this airspace w is at general risk.
[0136] If R z2 (t) <R z (w,t) <R z3 (t), this airspace w is at medium risk.
[0137] If R z (w,t)>R z3 (t), this airspace w is at significant risk.
[0138] Thus, the airspace risk rating of airspace w in the time interval from time t to t+1 is completed. The airspace risk quantification value R is calculated for all airspaces in the aviation network in all time intervals within the preset time period. z (w, t), and perform risk rating to form the risk evolution status of all airspaces over time.
[0139] Compared with the prior art, the aviation network risk assessment method provided by the present invention has the following advantages:
[0140] 1. Based on the risks of a single airport, studying the more serious airspace risks caused by risk cascades and conducting airspace-oriented risk assessments will help control the overall operating situation of the airspace.
[0141] 2. Potential safety risks such as airport saturation and flight delays are selected as assessment factors. Starting from the factors that cause aviation network risks in the pre-tactical stage, the time points for risk assessment and adjustment measures are advanced to prevent them from developing into major flight accidents such as collisions, thereby reducing high-risk operations at the tactical level.
[0142] 3. Flight delays and airport saturation are dynamic, which makes the aviation network security risk assessment based on delays and congestion dynamic, and the risk level assessment of each airspace in the entire aviation network more reliable.
[0143] The above is a detailed description of the aviation network risk assessment method provided by the present invention. For those skilled in the art, any obvious changes made to the invention without departing from the essence of the invention will fall within the scope of protection of the patent right of the present invention.
Claims
1. An aviation network risk assessment method, characterized in that The steps include: Get the average takeoff delay time and airport saturation of each airport in all time intervals within a specific time period; After normalizing the average takeoff delay time and the airport saturation of all airports in a certain time interval, clustering is performed to obtain m takeoff delay intervals and saturation intervals and m-1 takeoff delay time cutoff values and airport saturation cutoff values, where m is a positive integer; The average takeoff delay time and the airport saturation of each airport in a certain time interval are matched to the corresponding takeoff delay interval and saturation interval, and are respectively mapped into corresponding delay weight vectors and saturation weight vectors through corresponding mapping functions based on Gaussian kernel functions; wherein the delay weight vector is: Where Wd(i,t) is the delay weight vector of airport i in the time interval from t to t+1, d i (t) is the average takeoff delay time of airport i in the time interval from t to t+1, k is the takeoff delay interval number in the time interval from t to t+1, j is all the takeoff delay intervals before the current takeoff delay interval k; n is the number of takeoff delay intervals in the time interval from t to t+1 obtained by clustering; The saturation weight vector is: Among them, W s (i, t) is the saturation weight vector of airport i in the time period from t to t+1; s i (t) is the saturation value of airport i in the time interval from t to t+1, h is the saturation interval number in the time interval from t to t+1, v is all saturation intervals before the current saturation interval h; n is the number of saturation intervals in the time interval from t to t+1 obtained by clustering; According to the delay weight vector and the saturation weight vector of each airport in all time intervals, the risk quantification coefficient of each airport in all time intervals is obtained to form a corresponding risk time series of each airport; wherein the corresponding risk time series of each airport includes the corresponding risk quantification coefficient of all time intervals of each airport in a specific time period; According to the given airport risk quantification coefficient threshold, the risk airport set and non-risk airport set corresponding to each time interval are divided; According to the risk airport set corresponding to a certain time interval and the route connection relationship between airports, the risk airport geographical cluster of the aviation network in the time interval is obtained; According to the airspace range, the risk airport geographical clusters and the risk quantification coefficients of the airports within a certain time interval are cut to obtain the risk airport clusters in each airspace within the time interval; According to the geographical cluster of risk airports in each airspace within a certain time interval, the airport risk and cascade risk of each airspace in the time interval are obtained; According to the cascade risk and the airport risk of each airspace within a certain time interval, a quantitative value of airspace risk of each airspace is obtained; Obtain the airspace risk threshold within a certain time interval and divide the airspace risk level; Based on the airspace risk level in a certain time interval, a risk rating is performed on each airspace in the time interval.
2. The aviation network risk assessment method according to claim 1, characterized in that The average take-off delay time for all time intervals at each airport in a specific time period is obtained according to the following formula: Among them, d i (t) is the average takeoff delay time of airport i in the time interval from t to t+1, Di(t) is the total takeoff delay time of airport i in the time interval from t to t+1; Ci(t) is the number of cancelled flights at airport i in the time interval from t to t+1; 3 is the equivalent takeoff delay time of cancelled flights; P i (t) is the total number of scheduled departure flights from airport i in the time interval from t to t+1; t is a positive integer, representing the tth time interval.
3. The aviation network risk assessment method according to claim 1, characterized in that: The airport saturation of each airport in all time intervals within a specific time period is: the ratio of the arrival flow of airport i to the number of aircraft entering airport i in the time interval from t to t+1.
4. The aviation network risk assessment method according to claim 1, characterized in that: The corresponding risk quantification coefficient of all time intervals of each airport in a specific time period is the modulus of a two-dimensional plane vector synthesized by the delay weight vector and the saturation weight vector of the airport in the corresponding time interval.
5. The aviation network risk assessment method according to claim 1, characterized in that The formation process of the risk airport geographical cluster of the aviation network within a certain time interval includes the following steps: A risk airport is randomly selected from the risk airport cluster in a certain time interval as the starting airport of the risk airport geographical cluster, and related airports of the starting airport in the time interval are determined; Check whether the related airports of the selected risk airport are risk airports. If so, they are incorporated into the risk airport geographical cluster of the selected time interval. Otherwise, they are defined as isolated airports. Check whether the related airports of each risk airport incorporated into the risk airport geographical cluster are risk airports. If so, merge them into the risk airport geographical cluster until all the related airports of all risk airports newly incorporated into the risk airport geographical cluster are quarantine airports, and then obtain a risk airport geographical cluster for the selected time interval; The risk airports in the risk airport concentration of the selected time interval that do not belong to the determined risk airport geographical clusters will continue to be the starting airports of the next risk airport geographical cluster, and their related airports will be confirmed and it will be determined whether the related airports are dangerous airports, until all the airports in the risk airport concentration that can form risk airport geographical clusters are incorporated into the corresponding risk airport geographical clusters.
6. The aviation network risk assessment method according to claim 1, characterized in that The airport risk of each airspace in a certain time interval is obtained by taking the weighted average of the risk quantification coefficient values of the risk airports in each airspace; The cascade risk of each airspace in a certain time interval is obtained according to the following formula: Among them, R l (w,t) is the cascade risk of a certain space w in the time interval from t to t+1, l j(w,t) ,j=0,…,q w is the number of risk airports in the jth risk airport geographical cluster in airspace w, m w is the total number of airports in airspace w, q w is the number of risk airport geographical clusters within the airspace w.
7. The aviation network risk assessment method according to claim 1, characterized in that The quantitative value of airspace risk of a certain airspace in a certain time interval is obtained according to the following formula: Where α = m w / s, is the airport density of airspace w, m w is the total number of airports in airspace w, s is the total number of airports in the entire aviation network, β is the route density of airspace w and u w is the total number of routes in airspace w, R c (w,t) is the airport risk of a certain airspace w in the time interval from t to t+1, R l (w,t) is the cascading risk of the spatial domain w in the time interval from t to t+1.
8. The aviation network risk assessment method according to claim 1, characterized in that The process of obtaining the airspace risk threshold within a certain time interval is: The airport risk threshold and the airspace cascade risk threshold within a certain time interval are obtained respectively; According to the airport risk threshold and the airspace cascade risk threshold in a certain time interval, the airspace risk threshold of the time interval is obtained, and the airspace risk level is divided.
9. The aviation network risk assessment method according to claim 8, characterized in that The process of obtaining the airport risk threshold within a certain time interval is: Respectively obtaining a threshold value of the average take-off delay time and a threshold value of the airport saturation within a certain time interval; The threshold of the average take-off delay time and the threshold of the airport saturation in a certain time interval are corresponded to the corresponding take-off delay interval and saturation interval, and are respectively mapped into corresponding delay weight threshold vectors and saturation weight threshold vectors through corresponding mapping functions based on Gaussian kernel functions; According to the delay weight threshold vector and the saturation weight threshold vector in a certain time interval, an airport risk threshold in the time interval is obtained.
10. The aviation network risk assessment method according to claim 8, characterized in that When obtaining the airspace cascade risk threshold within a certain time interval, first obtain the airspace cascade risk value set of each airspace in different time intervals within the past preset time, and then perform clustering to obtain the airspace cascade risk threshold for a certain time interval.
