Risk prediction and optimization algorithm in intelligent civil aviation safety management system

Abstract

The application discloses a risk prediction and optimization algorithm in an intelligent civil aviation safety management system, relates to the technical field of risk control, and is used for improving the problem of low module adaptability caused by non-planned maintenance operation and leading to safety hazards, and comprises the following steps: self-inspection and judgment on whether intermittent module operation replacement occurs in the aircraft before and after flight operation, selection of whether to enter a scoring mechanism or a maintenance monitoring mechanism according to a judgment result, monitoring of behaviors of each module in the aircraft when the maintenance monitoring mechanism is entered, identification of a module locking state, explanation of a state change amplitude through a verification mechanism, calling of aircraft log to obtain aircraft maintenance data, combination of the state change amplitude to judge whether the aircraft is subjected to non-planned maintenance operation, acquisition of aircraft flight operation data and module change data, risk classification of the current non-planned maintenance operation, judgment on whether to enter a maintenance interception management system according to a risk classification result, screening of sensitive modules, and judgment on whether to perform alarm processing.

Description

Risk prediction and optimization algorithms in intelligent civil aviation safety management systems Technical Field

[0001] This invention relates to the field of risk management technology, and more specifically, to risk prediction and optimization algorithms in intelligent civil aviation safety management systems. Background Technology

[0002] Risk management technology is a systematic approach to identifying, assessing, and responding to potential risks. When applied to intelligent civil aviation safety management systems, risk management technology can effectively reduce management losses in civil aviation and improve the resilience and responsiveness of civil aviation safety management systems.

[0003] The existing technology has the following shortcomings:

[0004] In the past, when civil aircraft were conducting continuous flight operations, the parking area was protected by the safety control system of the parking area. However, this did not take into account the incompatibility issues caused by unplanned maintenance work performed on the aircraft due to hardware problems during flight operations or while parking. This greatly increased the flight risk when the aircraft was performing flight operations. Summary of the Invention

[0005] To overcome the aforementioned deficiencies in the prior art, embodiments of the present invention provide a risk prediction and optimization algorithm in an intelligent civil aviation safety management system. This algorithm performs self-checks on each module model after the aircraft completes flight operations and monitors the status changes of each module during parking, thereby screening sensitive modules and handling alarms to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The risk prediction and optimization algorithm in the intelligent civil aviation safety management system includes the following steps:

[0008] Step S1: When the aircraft completes the flight operation, check whether the model of each intermittent module in the aircraft is consistent with that before the current flight operation, determine whether the aircraft has been replaced during the intermittent module operation, and select to enter the replacement scoring mechanism or the maintenance monitoring mechanism based on the judgment result.

[0009] Step S2: When entering the maintenance monitoring mechanism, monitor the module behavior, identify the module lock status, interpret the state change range through the verification mechanism, call the machine log to obtain machine maintenance data, and determine whether the machine is performing unplanned maintenance work based on the state change range.

[0010] Step S3: When performing unplanned maintenance operations, access the aircraft action table to obtain the aircraft flight operation data within a set time, collect module change data, and classify risks by combining the flight operation data and module change data.

[0011] Step S4: Determine whether to enter the maintenance interception management system based on the risk classification results, and filter sensitive modules and determine whether to issue an alarm based on the output results of the maintenance interception management system.

[0012] In a preferred embodiment, in step S1, before the aircraft performs flight operations, the model of the intermittent module is recorded. The intermittent module includes a fuel injection module, a hydraulic system controller, and a climate control module.

[0013] When the aircraft completes a flight operation, the system detects and records whether the model of each intermittent module inside the aircraft is consistent with that before the current flight operation. If the model of each intermittent module is consistent in the two records, the status value is set to 0; if the model of each intermittent module is inconsistent in the two records, the status value is set to 1.

[0014] When the status value is 0, the machine enters the maintenance monitoring mechanism; when the status value is 1, the machine enters the replacement scoring mechanism.

[0015] In a preferred embodiment, in step S1, when the machine enters the replacement scoring mechanism, the currently replaced module model is called, a past period is selected as the analysis time, and the failure rate and maintenance frequency of the corresponding module model are collected during the analysis time. After standardization, the score of the corresponding module model is calculated using a linear regression algorithm, as follows: S = aF × bM + c, where S is the score of the corresponding module model, F is the failure rate of the corresponding module model, M is the maintenance frequency of the corresponding module model, a and b are the weight values ​​of F and M respectively, and c is a constant term used to adjust the scoring range; when the calculated score of the corresponding module model is lower than the preset scoring threshold, an alarm is triggered.

[0016] In a preferred embodiment, in step S2, the module behavior includes the distance of the module deviating from the default position within the machine body and the model change. The module locking state is set to three states: normal, deviation, and switching.

[0017] When the machine enters the maintenance monitoring mechanism, the behavior of the modules is monitored in real time, and the magnitude of the status change is explained through the set module lock status and verification mechanism.

[0018] In a preferred embodiment, in step S2, the rules for interpreting the state change magnitude are as follows:

[0019] If the module model does not change or the module does not deviate from its default position during real-time monitoring, the module's status is considered normal. If the module deviates from its default position during real-time monitoring, the module's status is considered deviated. If the module model changes during real-time monitoring, the module's status is considered switched. If both the module model and the module deviate from its default position during real-time monitoring, the module's status is considered both deviated and switched.

[0020] When the module status is "deviation", the ratio of the module's deviation from the default position to the comparison distance is used as the change value of the corresponding module; when the module status is "switch", the change value of the corresponding module is set to 1; when the module status is both "deviation" and "switch", the ratio of the module's deviation from the default position to the comparison distance plus 1 is used as the change value of the corresponding module.

[0021] In a preferred embodiment, in step S2, the machine body maintenance data includes the pressure values ​​of each module within the machine body and their location within the functional system. The mechanism for determining whether the machine body is undergoing unplanned maintenance operations, combined with the magnitude of the status change, is as follows:

[0022] Modules located in the same functional system are marked with the same label and the average pressure is calculated as the corresponding label pressure threshold. When the change value of each labeled module exceeds 1 and the pressure value exceeds the corresponding label pressure threshold, it is determined that the machine is performing unplanned maintenance work.

[0023] In a preferred embodiment, in step S3, the aircraft flight operation data is the aircraft flight mileage. A period of time is selected as the investigation time. During the investigation time, the flight mileage of all aircraft flight operations is accumulated to obtain the target flight distance of the aircraft.

[0024] The module change data consists of the runtime of each module during the aircraft's last flight operation. The risk coefficient of the aircraft is calculated using logistic regression, taking into account the target flight distance and the runtime of modules in abnormal states. The specific steps are as follows:

[0025] After normalizing the target flight distance and the runtime of the abnormal state modules within the aircraft, a logistic regression model is constructed: L = 1 / 1 + e -z Where L is the risk coefficient of the aircraft, e is the natural base, and z is the target flight distance of the aircraft and the weighted sum of the normalized results of the runtime of each abnormal state module in the aircraft.

[0026] When the risk coefficient of the machine exceeds the preset risk threshold, the current unplanned maintenance operation is judged as a high-risk maintenance operation; when the risk coefficient of the machine is lower than the preset risk threshold, the current unplanned maintenance operation is judged as a low-risk maintenance operation.

[0027] In a preferred embodiment, in step S4, if the current unplanned maintenance operation is determined to be a high-risk maintenance, the maintenance interception management system is entered. The maintenance interception management system detects each abnormal state module in the machine body, examines the operating condition range of each module, and obtains the operating condition data of each abnormal state module. The operating condition data includes a variety of data, including the operating temperature and humidity range, pressure range, and speed range of the module.

[0028] In a preferred embodiment, in step S4, the operational sensitivity coverage range of each abnormal state module is calculated based on the operational condition data, as detailed below:

[0029] The upper and lower intervals of all data within the module's operating condition data are standardized, and then summed and averaged to obtain the corresponding module's top and bottom data values.

[0030] The difference between the top and bottom values ​​of the data of a module is used to obtain the operating sensitive coverage range of the corresponding module. The average operating sensitive coverage range of all modules in the computer is used as the sensitivity judgment threshold. When the operating sensitive coverage range of an abnormal module is lower than the sensitivity judgment threshold, the abnormal module is judged to be a sensitive module.

[0031] The number of sensitive alarms is set to n. When unplanned maintenance work is performed, sensitive modules are filtered out by the maintenance interception management system and sent to the management port. When the number of sensitive modules in the abnormal state exceeds the number of sensitive alarms, alarm processing is performed.

[0032] The technical effects and advantages of the risk prediction and optimization algorithm in the intelligent civil aviation safety management system of this invention:

[0033] This invention determines whether intermittent modules within the aircraft have been replaced during flight operations by checking their model consistency before and after flight operations. Based on the determination, it selects between a scoring mechanism and a maintenance monitoring mechanism. The scoring mechanism performs self-checks to avoid safety hazards caused by intermittent modules during flight operations. The maintenance monitoring mechanism monitors the behavior of each module, identifies module lockout states, interprets the magnitude of state changes through a verification mechanism, retrieves maintenance data from the aircraft log, and determines whether unplanned maintenance is being performed based on the magnitude of state changes. When unplanned maintenance is being performed, flight operation data and module change data are acquired, and the current unplanned maintenance is risk-classified. This risk classification determines sensitive screening directions and detection costs. Based on the risk classification results, it determines whether to enter the maintenance interception management system. The system's output filters sensitive modules and determines whether to issue alarms, thereby improving the safety of the aircraft during subsequent flight operations. Attached Figure Description

[0034] Figure 1 is a schematic diagram of the risk prediction and optimization algorithm in the intelligent civil aviation safety management system of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] This invention improves the safety of the aircraft during its next flight operation by checking the consistency of the models of various intermittent modules before and after flight operations. Based on the results, it selects between a scoring mechanism and a maintenance monitoring mechanism. When the maintenance monitoring mechanism is activated, it monitors the behavior of each module, identifies module lockout states, interprets the magnitude of state changes through a verification mechanism, retrieves maintenance data from the aircraft log, and determines whether unplanned maintenance is being performed based on the magnitude of state changes. During unplanned maintenance, it acquires flight operation data and module change data, classifies the current unplanned maintenance as a risk, and determines whether to enter the maintenance interception management system based on the risk classification results. The system then filters sensitive modules and determines whether to issue an alarm based on the output of the maintenance interception management system, thereby enhancing the safety of the aircraft during its next flight operation.

[0037] An example of a risk prediction and optimization algorithm in an intelligent civil aviation safety management system, as shown in Figure 1, includes the following steps:

[0038] Step S1: When the aircraft completes the flight operation, check whether the model of each intermittent module in the aircraft is consistent with that before the current flight operation, determine whether the aircraft has been replaced during the intermittent module operation, and select to enter the replacement scoring mechanism or the maintenance monitoring mechanism based on the judgment result.

[0039] Step S2: When entering the maintenance monitoring mechanism, monitor the module behavior, identify the module lock status, interpret the state change range through the verification mechanism, call the machine log to obtain machine maintenance data, and determine whether the machine is performing unplanned maintenance work based on the state change range.

[0040] Step S3: When performing unplanned maintenance operations, access the aircraft action table to obtain the aircraft flight operation data within a set time, collect module change data, and classify risks by combining the flight operation data and module change data.

[0041] Step S4: Determine whether to enter the maintenance interception management system based on the risk classification results, and filter sensitive modules and determine whether to issue an alarm based on the output results of the maintenance interception management system.

[0042] The specific implementation is as follows:

[0043] In step S1, before the aircraft performs flight operations, the model of the intermittent module is recorded. The intermittent module includes the fuel injection module, the hydraulic system controller, and the climate control module.

[0044] The fuel injection module is used to precisely inject fuel into the combustion chamber of the engine, adjusting the amount of fuel injected according to the engine's operating status and demand.

[0045] The hydraulic system controller is used to manage the aircraft's hydraulic system, including the control surfaces, landing gear, and braking system;

[0046] The climate control unit is used to regulate the temperature and humidity inside the cabin, and maintain the air quality inside the cabin through heating, cooling and ventilation.

[0047] When the aircraft completes a flight operation, the system detects and records whether the model of each intermittent module inside the aircraft is consistent with that before the current flight operation. If the model of each intermittent module is consistent in the two records, the status value is set to 0; if the model of each intermittent module is inconsistent in the two records, the status value is set to 1.

[0048] When the status value is 0, the machine enters the maintenance monitoring mechanism; when the status value is 1, the machine enters the replacement scoring mechanism.

[0049] It should be explained that if the results of the two records for each intermittent module model are consistent, it means that the intermittent modules did not change during the flight operation, and the intermittent module models remained consistent before and after the flight operation without being replaced. Therefore, it is not necessary to enter the replacement scoring mechanism.

[0050] When the system enters the replacement scoring mechanism, it calls the currently replaced module model, selects a past period as the analysis time, and collects the failure rate and maintenance frequency of the corresponding module model during the analysis time. After standardization, it calculates the score of the corresponding module model using a linear regression algorithm, as follows: S = aF × bM + c, where S is the score of the corresponding module model, F is the failure rate of the corresponding module model, M is the maintenance frequency of the corresponding module model, a and b are the weight values ​​of F and M respectively, and c is a constant term used to adjust the scoring range. When the calculated score of the corresponding module model is lower than the preset scoring threshold, an alarm is triggered.

[0051] It should be noted that intermittent modules are airframe modules that can be intermittently deactivated during flight operations. This example only uses the fuel injection module, hydraulic system controller, and climate control module as examples. In reality, intermittent modules within the airframe are not limited to these three types. The constant term c in the linear regression algorithm is set by professionals in the field. The two weight values ​​corresponding to the standardized results of the failure rate and maintenance frequency of the module model are not unique and can be set according to the actual situation. For example, a can be set to 0.8 and b to 0.6, etc., which will not be elaborated here.

[0052] In step S2, the module behavior includes the distance of the module deviating from the default position within the machine body and the model change. The module locking state is set to three states: normal, deviation, and switching.

[0053] It should be explained that deviation and switching are two abnormal states of the module. In addition, abnormal states can be superimposed. In this example, the module can be in both deviation and switching states at the same time.

[0054] When the machine enters the maintenance monitoring mechanism, the behavior of the modules is monitored in real time. The magnitude of the status change is interpreted through the set module lock status and verification mechanism. The interpretation rules are as follows:

[0055] If the module model does not change or the module does not deviate from its default position during real-time monitoring, the module's status is considered normal. If the module deviates from its default position during real-time monitoring, the module's status is considered deviated. If the module model changes during real-time monitoring, the module's status is considered switched. If both the module model and the module deviate from its default position during real-time monitoring, the module's status is considered both deviated and switched.

[0056] It should be added that after judging the module, different colors can be selected to mark each module according to the judgment result. For example, when the module status is normal, it is marked as green; when the module status deviates, it is marked as yellow; when the module status is switched, it is marked as red. Furthermore, when the module status deviates or switches, it is marked as black. We will not go into too much detail here. Using different colors to mark each module facilitates the subsequent tracking and processing of each module.

[0057] When the module status is "deviation", the ratio of the module's deviation from the default position to the comparison distance is used as the change value of the corresponding module; when the module status is "switch", the change value of the corresponding module is set to 1; when the module status is both "deviation" and "switch", the ratio of the module's deviation from the default position to the comparison distance plus 1 is used as the change value of the corresponding module.

[0058] It should be noted that the comparison distance is a preset deviation from the comparison value, and its setting is not unique. For example, the comparison distance can be set to 1cm, etc., which will not be elaborated here.

[0059] The machine body maintenance data includes the pressure values ​​of each module within the machine body and their location within the functional system. The mechanism for determining whether unplanned maintenance work has been performed, based on the magnitude of changes in condition, is as follows:

[0060] Modules located in the same functional system are marked with the same label and the average pressure is calculated as the corresponding label pressure threshold. When the change value of each labeled module exceeds 1 and the pressure value exceeds the corresponding label pressure threshold, it is determined that the machine is performing unplanned maintenance work.

[0061] It should be noted that the above-mentioned judgment on non-calculated maintenance operations is carried out in real time after the aircraft completes the last flight operation. By monitoring the aircraft after the replacement and detection of intermittent module operations during the completion of flight operations, the behavior of the monitoring modules is used to determine whether the aircraft is performing unplanned maintenance operations, thereby reducing the safety hazards caused by unplanned maintenance operations. The aircraft log is an information document in aviation maintenance and operation management. In this example, it is used to obtain the pressure values ​​of each module in the aircraft and the location of its functional system.

[0062] In step S3, the aircraft flight operation data is the aircraft flight mileage. When it is determined that the aircraft is performing unplanned maintenance operations, the aircraft action table is accessed, a period of time is selected as the investigation time, and the flight mileage of all aircraft flight operations is accumulated during the investigation time to obtain the target flight distance of the aircraft.

[0063] The module change data consists of the runtime of each module during the aircraft's last flight operation. The risk coefficient of the aircraft is calculated using logistic regression, taking into account the target flight distance and the runtime of modules in abnormal states. The specific steps are as follows:

[0064] After normalizing the target flight distance and the runtime of the abnormal state modules within the aircraft, a logistic regression model is constructed: L = 1 / 1 + e -z Where L is the risk coefficient of the aircraft, e is the natural base, and z is the weighted sum of the normalized results of the target flight distance of the aircraft and the runtime of each abnormal state module in the aircraft.

[0065] When the risk coefficient of the machine exceeds the preset risk threshold, the current unplanned maintenance operation is judged as a high-risk maintenance operation; when the risk coefficient of the machine is lower than the preset risk threshold, the current unplanned maintenance operation is judged as a low-risk maintenance operation.

[0066] It should be noted that the greater the target flight distance of the aircraft, the more abnormal state modules there are, or the longer the flight operation time in the abnormal state modules, the greater the risk factor of the aircraft, and the more likely the aircraft is to experience abnormalities during flight operations.

[0067] In step S4, if the current unplanned maintenance operation is determined to be a high-risk maintenance, the system enters the maintenance interception management system. The maintenance interception management system detects each abnormal state module within the machine body, examines the operating condition range of each module, and obtains the operating condition data of each abnormal state module. The operating condition data includes various data, including the module's operating temperature and humidity range, pressure range, and speed range. Based on the operating condition data, the sensitive coverage range of each abnormal state module is calculated, as shown below:

[0068] The upper and lower intervals of all data within the module's operating condition data are standardized, and then summed and averaged to obtain the corresponding module's top and bottom data values.

[0069] For example, select the pressure range in the operating condition data of the module, take the logarithm of the upper and lower intervals of the pressure range to standardize it. Similarly, after processing all the operating condition data, take the weighted average of the upper intervals of various types of operating condition data as the top value of the corresponding module, and take the weighted average of the lower intervals of various types of operating condition data as the bottom value of the corresponding module.

[0070] The difference between the top and bottom values ​​of the data of a module is used to obtain the operating sensitive coverage range of the corresponding module. The average operating sensitive coverage range of all modules in the computer is used as the sensitivity judgment threshold. When the operating sensitive coverage range of an abnormal module is lower than the sensitivity judgment threshold, the abnormal module is judged to be a sensitive module.

[0071] The number of sensitive alarms is set to n. When unplanned maintenance work is performed, sensitive modules are filtered out by the maintenance interception management system and sent to the management port. When the number of sensitive modules in the abnormal state exceeds the number of sensitive alarms, alarm processing is performed.

[0072] It should be noted that filtering out sensitive modules makes it easier for administrators to quickly handle changes to the modules. Filtering out sensitive modules also facilitates subsequent management and maintenance of these modules. At the same time, by setting the number of sensitive alarms, alarms can be triggered for unplanned maintenance operations, reducing the safety hazards that unplanned maintenance operations may pose to the machine.

[0073] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0074] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0075] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0076] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0077] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A risk prediction and optimization algorithm in an intelligent civil aviation safety management system, characterized in that, The process includes the following steps: Step S1: When the aircraft completes a flight operation, check whether the model of each intermittent module within the aircraft is consistent with that before the current flight operation, determine whether the aircraft has undergone replacement during the intermittent module operation, and select to enter the replacement scoring mechanism or the maintenance monitoring mechanism based on the judgment result; Step S2: When entering the maintenance monitoring mechanism, monitor the module behavior, identify the module lock status, interpret the state change range through the verification mechanism, call the aircraft log to obtain the aircraft maintenance data, and determine whether the aircraft has performed unplanned maintenance operations based on the state change range; Step S3: When performing unplanned maintenance operations, access the aircraft action table to obtain the aircraft flight operation data within a set time, collect module change data, and classify risks by combining the flight operation data and module change data; Step S4: Based on the risk classification result, determine whether to enter the maintenance interception management system, filter sensitive modules and determine whether to issue an alarm based on the output result of the maintenance interception management system; In Step S2, module behavior includes the distance of the module within the aircraft from the default position and model change, and the module lock status is set to three states, divided into The states are: normal, deviation, and switching. When the machine enters the maintenance monitoring mechanism, the module behavior is monitored in real time, and the state change range is interpreted through the set module lock status and verification mechanism. In step S2, the rules for interpreting the state change range are as follows: If the machine does not change the module model or the module does not deviate from the default position during real-time monitoring, the module is judged to be in normal state. If the machine only deviates from the default position during real-time monitoring, the module is judged to be in deviation state. If the machine only changes the module model during real-time monitoring, the module is judged to be in switching state. If the machine changes the module model and deviates from the default position simultaneously during real-time monitoring, the module is judged to be in both deviation and switching state. When the module state is deviation, the ratio of the distance of the module from the default position to the comparison distance is used as the change range value of the corresponding module. When the module state is switching, the change range value of the corresponding module is set to 1. When the module state is both deviation and switching, the ratio of the distance of the module from the default position to the comparison distance plus 1 is used as the change range value of the corresponding module.

2. The risk prediction and optimization algorithm in the intelligent civil aviation safety management system according to claim 1, characterized in that: In step S1, before the aircraft performs flight operations, the model numbers of intermittent modules are recorded. Intermittent modules include fuel injection modules, hydraulic system controllers, and climate control modules. When the aircraft completes flight operations, the model numbers of each intermittent module inside the aircraft are checked and recorded to see if they are consistent with those before the current flight operation. If the model numbers of each intermittent module are consistent in the two records, the status value is set to 0. If the model numbers of each intermittent module are inconsistent in the two records, the status value is set to 1. When the status value is 0, the aircraft enters the maintenance monitoring mechanism. When the status value is 1, the aircraft enters the replacement scoring mechanism.

3. The risk prediction and optimization algorithm in the intelligent civil aviation safety management system according to claim 1, characterized in that: In step S1, when the machine enters the replacement scoring mechanism, the currently replaced module model is called, a past period is selected as the analysis time, and the failure rate and maintenance frequency of the corresponding module model are collected during the analysis time. After standardization, the score of the corresponding module model is calculated using a linear regression algorithm, as follows: Where S is the score of the corresponding model module, F is the failure rate of the corresponding model module, M is the maintenance frequency of the corresponding model module, a and b are the weight values ​​of F and M respectively, and c is a constant term used to adjust the scoring range; when the calculated score of the corresponding model module is lower than the preset scoring threshold, an alarm is triggered.

4. The risk prediction and optimization algorithm in the intelligent civil aviation safety management system according to claim 1, characterized in that: In step S2, the machine body maintenance data includes the pressure values ​​of each module within the machine body and their location in the functional system. The mechanism for determining whether the machine body is undergoing unplanned maintenance work is as follows: Modules located in the same functional system are marked with the same label and the average pressure is calculated as the corresponding label pressure threshold. When the change value of each labeled module exceeds 1 and the pressure value exceeds the corresponding label pressure threshold, it is determined that the machine body is undergoing unplanned maintenance work.

5. The risk prediction and optimization algorithm in the intelligent civil aviation safety management system according to claim 1, characterized in that: In step S3, the aircraft flight operation data is the aircraft's flight mileage. A period of time is selected as the investigation time, and the flight mileage of all aircraft flight operations is accumulated within the investigation time to obtain the aircraft's target flight distance. The module change data is the runtime of each module during the aircraft's last flight operation. Combining the aircraft's target flight distance and the runtime of abnormal state modules within the aircraft, the logistic regression method is used to calculate the aircraft's risk coefficient. The specific steps are as follows: After normalizing the aircraft's target flight distance and the runtime of abnormal state modules within the aircraft, a logistic regression model is constructed: Where L is the risk coefficient of the aircraft, e is the natural base, and z is the weighted sum of the target flight distance of the aircraft and the normalized running time of each abnormal state module in the aircraft. When the risk coefficient of the aircraft exceeds the preset risk threshold, the current unplanned maintenance operation is judged as a high-risk maintenance operation. When the risk coefficient of the aircraft is lower than the preset risk threshold, the current unplanned maintenance operation is judged as a low-risk maintenance operation.

6. The risk prediction and optimization algorithm in the intelligent civil aviation safety management system according to claim 5, characterized in that: In step S4, if the current unplanned maintenance operation is judged to be a high-risk maintenance, the maintenance interception management system is entered. The maintenance interception management system detects each abnormal state module in the machine body, examines the operating condition range of each module, and obtains the operating condition data of each abnormal state module. The operating condition data includes a variety of data, including the operating temperature and humidity range, pressure range, and speed range of the module.

7. The risk prediction and optimization algorithm in the intelligent civil aviation safety management system according to claim 6, characterized in that: In step S4, the sensitive coverage range of each abnormal state module is calculated based on the operating condition data, as follows: After standardizing the upper and lower intervals of all data in the module's operating condition data, the summation and average value are obtained to get the top and bottom values ​​of the corresponding module's data; the difference between the top and bottom values ​​of the module's data is obtained to get the sensitive coverage range of the corresponding module's operation; the average value of the sensitive coverage range of all modules in the computer is used as the sensitivity judgment threshold; when the sensitive coverage range of an abnormal state module is lower than the sensitivity judgment threshold, the abnormal state module is judged to be a sensitive module. The number of sensitive alarms is set to n. When unplanned maintenance work is performed, sensitive modules are filtered out by the maintenance interception management system and sent to the management port. When the number of sensitive modules in the abnormal state exceeds the number of sensitive alarms, alarm processing is performed.

Images