Passenger cabin bumping event emergency decision-making method based on improved bow tie-continuous Bayesian network model

Abstract

The invention discloses a passenger cabin bumping event emergency decision-making method based on an improved bow-tie-continuous Bayesian network model, and the method comprises the following steps: determining a disposal node and a node relation according to a preset bumping event emergency disposal process; constructing a bow-tie model according to the disposal nodes and the node relationship, and recording node states under a plurality of time slices before and after a bumping event through the bow-tie model to obtain a dynamic bow-tie model; the dynamic bow tie model is converted into a continuous Bayesian network model, and the transition probability of any node under different time slices is calculated through a preset transition mode; according to the method, the continuous Bayesian network model is formed by improving the bow-tie model, simulation of the state of each node along with time change before and after the bumping event can be realized, and dynamic analysis of each node can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency decision-making, and more specifically to an emergency decision-making method for cabin turbulence events based on an improved bowtie-continuous Bayesian network model. Background Art

[0002] Turbulence refers to a meteorological disaster in which an aircraft is impacted by disturbed air currents during flight, causing inaccurate instruments and difficulty in operation.

[0003] At present, the research on the technology of turbulence warning before the event and the research on the methods of emergency response to sudden incidents after the event are concentrated at home and abroad: on the one hand, turbulence warning focuses on analyzing the causes of cabin turbulence from a meteorological perspective before the flight, and then developing technologies such as pre-flight turbulence prediction and warning. However, although the existing turbulence warning technology can identify most convective turbulence, it is difficult to achieve accurate warning for clear air turbulence. In the existing technology, cabin sudden incidents are regarded as static events, focusing on accident case reasoning and response strategies after a single event occurs, ignoring the impact of event levels and continuous response processes. On the other hand, the emergency response methods for sudden incidents after the event occur, focusing on personnel evacuation, emergency decision-making and event cause research after sudden incidents such as cabin turbulence and fire occur during flight. In the existing technology, there is a lack of targeted research on the turbulence event itself, especially for the inability to accurately warn sudden turbulence, the inability to dynamically monitor the specific situation after the turbulence occurs, and the key emergency strategies that consider dynamic changes are given. Summary of the invention

[0004] In view of this, the present invention provides an emergency decision-making method for cabin turbulence events based on the improved bowtie-continuous Bayesian network model, which takes into account the characteristics of such emergencies as unwarned, degree impact, and time changes. By constructing a dynamic evaluation model, the key nodes and emergency strategies of the events are scientifically analyzed to reduce the consequences of the events such as casualties, and can provide a methodological reference for similar emergencies in closed spaces, with difficult rescue and dynamic handling.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] An emergency decision-making method for cabin turbulence events based on an improved bowtie-continuous Bayesian network model comprises the following steps:

[0007] Confirm the disposal nodes and node relationships.

[0008] A bowtie model is constructed according to the handling nodes and the node relationships, and the node states at different moments in a turbulence event and the turbulence degree of the turbulence event are recorded through the bowtie model.

[0009] The bowtie model corresponding to different moments is converted into a continuous Bayesian network model under multiple time slices.

[0010] Initializing the initial probability of each node in the continuous Bayesian network model according to the degree of turbulence; calculating the transfer probability in adjacent time slices according to the initial probability, and obtaining the node probability distribution in each time slice in combination with the observed value; and confirming the key nodes according to the node probability distribution.

[0011] Preferably, the node probability distribution in each time slice obtained by combining the observations includes:

[0012] The joint transition probability in a certain time slice is calculated according to the node transition probability.

[0013] According to the joint transfer probability, various observation values ​​corresponding to the start and the time slice are introduced, and the node probability corresponding to each node is calculated to obtain the node probability distribution under the corresponding time slice.

[0014] Preferably, the initial probability is a priori probability or a posterior probability under the observed value.

[0015] When the initial probability is the prior probability, the node probability distribution is used to predict the dynamic changes of each node before and after the turbulence event occurs.

[0016] When the initial probability is the posterior probability under the observed value, the cause of the event is analyzed through the node probability.

[0017] Preferably, the step of obtaining the prior probability includes:

[0018] Obtain statistical data of events occurring at each node and calculate the first prior probability;

[0019] Obtain configuration data of events occurring at each node and calculate a second prior probability;

[0020] The first prior probability and the second prior probability are combined to determine a final prior probability.

[0021] Preferably, the transfer probability is calculated as follows:

[0022]

[0023] Among them, X t is the node set of the transfer network corresponding to time slice t, X t+Δt is the set of nodes on the time slice t+Δt; is the i-th node on the t-th time slice, For Node The parent node of , N is the total number of nodes in each time slice.

[0024] Preferably, the joint transition probability is calculated as follows:

[0025]

[0026] in, is the node state of node n in the corresponding transfer network at time slice t, and X0 represents the node set in the initial network.

[0027] Preferably, the bowtie model corresponding to different moments is converted into a continuous Bayesian network model under multiple time slices, specifically including:

[0028]

[0029] Among them, A d A represents the set of all nodes in the improved bowtie model, n is the set of parent nodes and top events, n∈(1,2…n); m is an intermediate event, m represents the number of the intermediate event, a z is the bottom event, z represents the bottom time number; A is the top event, and DBN consists of the initial network X0 and the dynamic network X t X0 represents the network set composed of the initial nodes at t=0; X t represents the network set composed of dynamic nodes at time slice t; X z (0) represents the root node, X m (0) represents the intermediate node; X (0) represents a leaf node; Δt∈(1,2…t) represents the time change.

[0030] Preferably, a bowtie model is constructed based on the handling nodes and the node relationships, including: taking the occurrence of turbulence as the top event A, and confirming various levels of events of event occurrence factors and various levels of events of event occurrence decisions based on the top event A.

[0031] Preferably, each level of events of the event occurrence decision includes multiple levels of turbulence representative events which are arranged in sequence according to the evolution order of the degree of turbulence after the turbulence occurs.

[0032] A cabin turbulence event emergency decision system based on an improved bowtie-continuous Bayesian network model comprises a computer system, wherein the computer system is used to implement the above emergency decision method.

[0033] It can be seen from the above technical solution that, compared with the prior art, the present invention discloses a method for emergency decision-making for cabin turbulence events based on an improved bowtie-continuous Bayesian network model. By analyzing the emergency response process of cabin turbulence events, an improved bowtie model reflecting the structured node relationship is constructed, and mapping conditions and conversion rules are established to form a continuous Bayesian network model. When constructing the model, the characteristics of such emergencies as turbulence, such as the lack of warning, degree influence, and time change, are taken into account to form a dynamic model, which is helpful to scientifically analyze the key nodes and emergency strategies of the events to reduce the consequences of events such as casualties, and can provide a method reference for similar emergencies in closed spaces, difficult rescue, and dynamic handling. At the same time, through the analysis of joint nodes, the non-critical nodes in the emergency response process can be indirectly reflected, which is also helpful for the optimization of the emergency response process. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0035] Figure 1 A schematic diagram of an emergency decision-making method for cabin turbulence events based on an improved bowtie-continuous Bayesian network model provided by the present invention.

[0036] Figure 2 is a schematic diagram of the structure of an improved bow tie model in an embodiment of the present invention;

[0037] Figure 3 A schematic diagram of a method for transforming an improved bowtie model into a continuous Bayesian network model in an embodiment of the present invention;

[0038] Figure 4 Schematic diagram of the structure of the converted Bayesian network model in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] 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.

[0040] Example 1

[0041] The embodiment of the present invention discloses an emergency decision-making method for cabin turbulence events based on an improved bowtie-continuous Bayesian network model, comprising the following steps:

[0042] S1: Confirm the disposal nodes and node relationships.

[0043] S2: Construct a bowtie model based on the disposal nodes and node relationships, and use the bowtie model to record the node status at different times in the turbulence event and the turbulence degree of the turbulence event.

[0044] S3: Convert the bowtie model corresponding to different moments into a continuous Bayesian network model under multiple time slices;

[0045] S4: Initialize the initial probability of each node in the continuous Bayesian network model according to the degree of turbulence; take the node status recorded by the bowtie model at different times as the observation value; calculate the node probability distribution at each time slice according to the initial probability and the observation value, and confirm the key nodes according to the node probability distribution.

[0046] In this embodiment, the present invention introduces time slices into the standard bow tie model, so that the bow tie model is more dynamic, so that the continuous Bayesian network model can be used to analyze the changes of the node status of each node over time before and after the bump. Among them, the continuous Bayesian network model includes an initial node network and a transfer node network. When the continuous Bayesian network model is used to dynamically analyze the bump event, the transfer network in the Bayesian network model analyzes the transfer probability of each node and realizes the network update of the node according to different node observation values, so as to clarify the dynamic impact of each node on the bump event.

[0047] For example, when the transfer probabilities of nodes A111 (captain misjudgment of turbulence level), A121 (crew fatigue flight) and A123 (crew failure to protect themselves) are known, it can be predicted that in the following period, if the probability of captain misjudgment increases, the probability of crew emergency response failure will also increase. If the probability of crew fatigue flight is high, the probability of crew failure to protect themselves will also increase accordingly.

[0048] In addition, based on the continuous Bayesian network model, the functions of forward reasoning and reverse reasoning can also be realized by combining prior probability or posterior probability. Specifically, the continuous Bayesian network for cabin turbulence emergency response is constructed using GeNle software, with prior probability P A Data analysis is performed by first performing forward reasoning to predict the dynamic changes in the emergency response time for cabin turbulence, helping decision makers take early intervention measures. Second, reverse reasoning is performed to obtain the posterior probability of each key node when the risk is at time t, analyze the key causes of turbulence events, and obtain a continuous Bayesian network for cabin turbulence emergency response.

[0049] Forward reasoning starts from known evidence, propagates forward along the structure of the Bayesian network, and gradually infers the probability distribution of other nodes. This reasoning method can be used to predict the state of unknown variables, or estimate the state of remaining variables when some variables are known.

[0050] In the process of forward reasoning, it is first necessary to initialize each node in the continuous Bayesian network model according to the prior probability of each node; wherein the prior probability refers to the initial probability distribution of the node under a certain degree of turbulence.

[0051] After obtaining the initial probability distribution, the transfer probability is calculated in combination with the initial probability distribution, and then the probability distribution of the current time slice is updated by introducing the observation value. This can obtain the dynamic changes of each node before and after the occurrence of the bumpy time.

[0052] Backward reasoning is the process of starting from the target node and propagating backward along the Bayesian network structure to gradually obtain the required information. This method of reasoning is usually used to find out the reasons that may lead to the result when the result is known.

[0053] In the process of reverse reasoning, it is necessary to give a posterior probability under a certain observation value, and then calculate the probability distribution of nodes before the observation value by combining the observation value to reversely reason the observation value, so as to determine the cause of the observation value state. Therefore, the key time slice and handling node can be obtained based on the occurrence of the turbulence event.

[0054] Next, each step in the present invention is described in detail.

[0055] For S1, in order to clarify the handling nodes and node relationships. It is necessary to clarify the emergency handling process of cabin turbulence incidents. In one embodiment, the turbulence incident can be divided into three stages according to the handling time: before the flight, during the flight, and after the flight. The key personnel for turbulence handling are divided into passengers, cabin crew, flight crew, and airline AOC personnel. In addition, in order to clarify the entire process of the cause of the cabin emergency handling during the flight, the handling nodes and node relationships of each aspect are confirmed from four aspects: weather conditions, mechanical equipment, safety supervision, and human factors.

[0056] like Figure 2For S2, when constructing the improved bowtie model, the four aspects of human-machine-environment-management are first selected to identify the overall risk factors, and the cabin turbulence causing casualties is taken as the top event. The intermediate events and basic events that lead to the top event are analyzed from top to bottom, and 23 key disposal nodes are selected to construct the accident tree. The top event is the occurrence of A turbulence event. The first-level intermediate events are: A1 human factors; A2 management factors; A3 environmental factors; A4 airborne equipment factors. The second level A11 Captain's emergency response failure; A12 Crew emergency response failure; A13 Passenger emergency failure; A21 internal reasons; A22 external reasons; A31 Bad weather environment; A32 The aircraft cannot be bypassed; A41 Fixture failure; A42 Weather radar failure. The third level is A5 pre-flight stage; A6 in-flight stage. The fourth level is A111 misjudgment of turbulence level; A112 the captain did not take proper measures to fix the aircraft; A113 did not broadcast the turbulence; A114 did not hold a tripartite meeting; A115 the captain was fatigued; A121 the crew was fatigued; A122 did not check the aircraft items to be fixed; A123 did not take proper self-protection measures; A124 failed to deal with the service; A125 did not fasten the seat belt; A132 was not in the seat; A133 did not fasten the seat belt; A211 the crew members were not trained properly; A212 the crew over-served; A213 the crew had psychological problems; A221 the regulations were incomplete; A222 the supervision was not strong enough. The fifth level is A1241 did not fix the dining car; A1242 did not fix themselves; A1321 did not hold the handrail tightly; A1322 did not fix themselves.

[0057] In addition, the occurrence of A turbulence is taken as the initial event. According to the time sequence of the turbulence events, the handling decisions are divided into 7 layers to draw an event tree based on the impact of the turbulence degree, the handling failure barrier and the severity of the event consequences. According to accident statistics and relevant laws and regulations, the degree of evolution after the turbulence is divided into seven levels, including: whether the turbulence stops, whether there are casualties, whether the casualties are serious, whether to make an alternate landing, whether the landing is smooth, whether the rescue is coordinated, and whether the event is coordinated. Finally, the improved bowtie model for cabin turbulence emergency handling, namely the dynamic bowtie model, is obtained.

[0058] For S3, the first step is to improve the transformation of the bowtie model to the continuous Bayesian model, such as Figure 3 The mapping transformation from the improved bowtie model to the continuous Bayesian network model needs to meet the following conditions: 1. The nodes are one-to-one corresponding, that is, a node in the bowtie model corresponds to a node in the Bayesian network model; the hierarchical relationship of each node in the bowtie model corresponds to the root-leaf node relationship in the Bayesian network model; 2. The transformation process has a certain time sequence, and according to the preset time slice interval Δt, the transformation is performed every Δt time slices.

[0059] Specifically, the mapping expression is:

[0060]

[0061] In formula (1), A d A represents the set of all nodes in the improved bowtie model, n is the set of parent nodes and top events, n∈(1,2…n). m is an intermediate event, m represents the number of the intermediate event, A z is the bottom event, z represents the bottom time number. A is the top event. DBN consists of the initial network X0 and the dynamic network X t X0 represents the network set composed of the initial nodes at t=0; X t represents the network set composed of dynamic nodes at time slice t; X z (0) represents the root node, X m (0) represents the intermediate node; X (0) represents a leaf node. Δt∈(1,2…t) represents the time change.

[0062] After the transformation, the relationship between each node in the Bayesian network model is as follows: Figure 4 As shown in the figure, the arrows in the figure point from the cause factor to the result factor, that is, the parent node points to the child node, where the node without a parent node is the root node, and the directed arc "1" on each root node indicates that the time slice interval is 1. The node status in DBN is described by appropriate qualitative language according to the name of the node, such as "yes, no".

[0063] Secondly, the prior probability is brought into the constructed node graph for further calculation.

[0064] Specifically, the transfer rules between different time slices of the continuous Bayesian network model: Since the continuous Bayesian transfer network of cabin turbulence is a Bayesian network model on multiple time slices, it satisfies the stationarity of the Markov assumption, and the cabin turbulence emergency response decision is only affected by the previous time slice, that is, the probability of the node at time t is only affected by the moment t-1, and has nothing to do with the time slice before t-1. Therefore, the conditional probability of the nodes in the network model converted from the improved bowtie model is exactly the same as that in the initial network, and the transfer probability remains unchanged throughout the continuous Bayesian network model. The transfer probability of any two adjacent time slices in the continuous Bayesian network is expressed as follows:

[0065]

[0066] In formula (2), after the improved bowtie model mapping, X t+Δt is the node set on the time slice t+Δt, X t is the set of nodes on the time slice t; is the value of the i-th node at the t+Δt-th time slice, For Node The parent node of a node can be located in the same or adjacent time slices. N is the total number of nodes in each time slice, N∈(1,2…n); nodes cannot be connected across time slices according to the Markov chain principle.

[0067] In order to further implement the above technical solution, according to the prior probability P of the initial network X0 A and the transition probability P(X t |X t-1 ), the joint transition probability of any node in the continuous Bayesian network is calculated as follows:

[0068]

[0069] In formula (3), It refers to the situation of node n in time slice t. is the transition probability of the initial network. When calculating, the parent node of the initial network adopts the prior probability P of the initial network. A ; is the transition probability of the transition network.

[0070] In this embodiment, the node probability distribution in any time slice can be calculated by combining the transition probability with the corresponding observation value.

[0071] In order to further implement the above technical solution, the step of obtaining the prior probability includes: obtaining the statistical data of the events occurring at each node and calculating the first prior probability; obtaining the configuration data of the events occurring at each node and calculating the second prior probability. The first prior probability and the second prior probability are the direct prior probability and the supplementary prior probability, respectively.

[0072] Specifically, the prior probability P required for continuous Bayesian calculation is A , which is composed of the direct prior probability and the supplementary prior probability which can be directly counted. The supplementary prior probability is obtained by the expert evaluation method of triangular fuzzy numbers.

[0073] Among them, the direct prior probability is the statistical probability P(C n )as follows:

[0074]

[0075] In formula (4), C a is the number of occurrences of the nth node under the statistical overview, P(C a ) represents the probability of the nth node occurring, C all It refers to the number of occurrences of all nodes, i = 1, 2, 3…n (n is the total number of events).

[0076] Supplementary node prior probability is fuzzy probability P G The calculation is as follows: First, eight senior experts in airline cabin management were invited to assign weights to the degree of experience based on industry relevance such as years of experience, flight time, and industry qualifications. Secondly, a questionnaire for moderate and severe turbulence was established based on the turbulence handling process. Each question was divided into five language variables based on the probability of occurrence from never occurring to always occurring for the experts to score. Finally, the fuzzy total G of each node was obtained through the questionnaire. i :

[0077]

[0078] In the formula, G i is the total number of fuzzy numbers, i=1,2,3…n, W i is the weight G of the i-th expert ij is the score of the jth event i, and B is the total number of experts. Substituting the obtained fuzzy total into formula (6), we can get the triangular fuzzy number G of the relevant node.

[0079]

[0080] In formula (6), G = (k, l, u) is a triangular fuzzy number, and its corresponding membership function μ G (A) indicates that k and u are the upper and lower limits of triangular fuzzy numbers respectively, and l is the value when the fuzzy number membership is 1.

[0081] According to the triangular fuzzy number obtained by statistical calculation, the final fuzzy probability is:

[0082]

[0083] To facilitate subsequent simulation, different weights can be assigned to statistical probability and fuzzy probability, such as statistical probability P(C n ) is assigned a weight of 0.7, and the prior probability P obtained by the triangular fuzzy number G Assign a weight of 0.3 and obtain all final prior probabilities.

[0084] It is understandable that, for the prior probability, in addition to taking the weighted sum of the statistical probability and the fuzzy probability as the final prior probability, the statistical probability or the fuzzy probability can also be selected as the final prior probability according to actual conditions.

[0085] In this embodiment, the present invention realizes the whole process supervision of cabin turbulence events by constructing an improved bowtie-continuous Bayesian network model, grasps the change process and trend of each node over time, calculates the node probability, performs data simulation and emulation, and analyzes the turbulence events.

[0086] Exemplarily, the prior probabilities of moderate turbulence and severe turbulence are used for initialization, and the initialization parameters are shown in the following table:

[0087]

[0088] According to the calculated prior probabilities of different nodes with moderate and severe turbulence, the node status is set according to the actual situation of the node. On this basis, dynamic indicators are selected to consider the time-varying characteristics of the node and set it as a dynamic node. The number of time slices is reasonably set using the GeNle software. The calculation logic of the GeNle software is the transition probability and joint probability mentioned above.

[0089] First, the dynamic Bayesian network for cabin turbulence emergency response was constructed using GeNle software, and the graph was brought in with prior probability for data analysis. Through data simulation, the probability distribution of each node was obtained. In the experiment, the change amplitude of the time slice every 2 minutes was more significant, so the node result analysis was performed every 2 time slices starting from t=1. When t≥8, the degree of change was almost the same as that at t=7 and was no longer significant. Therefore, the probability distribution results of each node under the change of moderate turbulence and severe turbulence at time slices of t=1, t=3, t=5, and t=7 were obtained.

[0090] (1) Emergency response is time-sensitive and there is an optimal response time. Under moderate turbulence, the probability of node failure increases with the increase of response time. Similarly, the change in the probability of node failure under severe turbulence shows the same trend as that under moderate turbulence. For example, at t = 1, the failure probability of severe turbulence at each node is higher than that of moderate turbulence, but as time increases, at t = 7, the failure probabilities of the two are almost equal. The failure probabilities of moderate and severe turbulence are almost the same at t = 5 and t = 7, and tend to be stable, indicating that the optimal emergency response time is before t = 5. (2) Emergency response strategies are affected by the degree of turbulence and have differences. At t = 1, in a horizontal comparison of the nodes of moderate and severe turbulence, the probability of crew members not taking good care of themselves A123, passengers and crew members not wearing seat belts A125 and A133, no fixed dining car A1241, and bad weather A31 changes significantly, with the change exceeding 0.2. This indicates that when the turbulence becomes more severe, inadequate personnel fixation measures are the main reason for the failure of emergency response.

[0091] (3) The probability of emergency response failure at each node caused by human factors is the highest. Among all the causal factors, A123, A1241, A1242, A133 and A212 are the key nodes that lead to the failure of handling cabin turbulence events. The failure probability of them in different time slices and different degrees of turbulence is at a high level. This shows that the failure of passengers to fasten their seat belts and the excessive service of cabin crew are the key reasons for the failure of handling.

[0092] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0093] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An emergency decision-making method for cabin turbulence events based on an improved bowtie-continuous Bayesian network model, characterized in that: The following steps are involved: Confirm disposal nodes and node relationships; Constructing a bowtie model according to the handling nodes and the node relationships, and recording the node states at different moments in a bumpy event and the bumpy degree of the bumpy event through the bowtie model; Convert the bowtie model corresponding to different moments into a continuous Bayesian network model under multiple time slices; Initializing the initial probability of each node in the continuous Bayesian network model according to the degree of turbulence; calculating the transfer probability in adjacent time slices according to the initial probability, and obtaining the node probability distribution in each time slice in combination with the observed value; and confirming the key nodes according to the node probability distribution.

2. The method for emergency decision-making of cabin turbulence events based on the improved bowtie-continuous Bayesian network model according to claim 1, characterized in that: The node probability distribution under each time slice is obtained by combining the observation values, including: Calculate the joint transition probability in a certain time slice according to the node transition probability; According to the joint transfer probability, various observation values ​​corresponding to the start and the time slice are introduced, and the node probability corresponding to each node is calculated to obtain the node probability distribution under the corresponding time slice.

3. The method for emergency decision-making of cabin turbulence events based on the improved bowtie-continuous Bayesian network model according to claim 1 or 2, characterized in that: The initial probability is a priori probability or a posterior probability under the observed value; When the initial probability is the prior probability, the dynamic changes of each node before and after the occurrence of the turbulence event are predicted by the node probability distribution; When the initial probability is the posterior probability under the observed value, the cause of the event is analyzed through the node probability.

4. The cabin incident emergency decision-making method based on the improved bowtie-continuous Bayesian network model according to claim 3 is characterized in that: The step of obtaining the prior probability comprises: Obtain statistical data of events occurring at each node and calculate the first prior probability; Obtain configuration data of events occurring at each node and calculate a second prior probability; The first prior probability and the second prior probability are combined to determine a final prior probability.

5. The method for emergency decision-making of cabin turbulence events based on the improved bowtie-continuous Bayesian network model according to claim 1, characterized in that: The calculation method of the transition probability is: Among them, X t is the node set of the transfer network corresponding to time slice t, X t+Δt is the set of nodes on the time slice t+Δt; is the i-th node on the t-th time slice, For Node The parent node of , N is the total number of nodes in each time slice.

6. A cabin event emergency decision-making method based on an improved bowtie-continuous Bayesian network model according to claim 2 or 5, characterized in that: The calculation method of the joint transition probability is: in, is the node state of node n in the corresponding transfer network at time slice t, and X0 represents the node set in the initial network.

7. The cabin incident emergency decision-making method based on the improved bowtie-continuous Bayesian network model according to claim 1, characterized in that: The converting the bowtie model corresponding to different moments into a continuous Bayesian network model under multiple time slices specifically includes: Among them, A d A represents the set of all nodes in the improved bowtie model, n is the set of parent nodes and top events, n∈(1,2…n); m is an intermediate event, m represents the number of the intermediate event, A z is the bottom event, z represents the bottom time number; A is the top event, and DBN consists of the initial network X0 and the dynamic network X t X0 represents the network set composed of the initial nodes at t=0; X t represents the network set composed of dynamic nodes at time slice t; X z (0) represents the root node, X m (0) represents the intermediate node; X (0) represents a leaf node; Δt∈(1,2…t) represents the time change.

8. The method for emergency decision-making of cabin turbulence events based on the improved bowtie-continuous Bayesian network model according to claim 1, characterized in that: A bowtie model is constructed according to the handling nodes and the node relationships, including: taking the occurrence of turbulence as the top event A, and confirming various levels of events of event occurrence factors and various levels of events of event occurrence decisions based on the top event A.

9. The method for emergency decision-making of cabin turbulence events based on the improved bowtie-continuous Bayesian network model according to claim 8, characterized in that: The various levels of events of the event occurrence decision include multiple levels of turbulence representative events that are sequentially arranged according to the order of evolution of the degree of turbulence after the turbulence occurs.

10. An emergency decision-making system for cabin turbulence events based on an improved bowtie-continuous Bayesian network model, characterized in that: It comprises a computer system, and the computer system is used to implement the emergency decision-making method described in any one of claims 1-9.

Images