Data driving and physical modeling fused landing gear buffer strut evaluation and prediction method

Abstract

The embodiment of the invention provides a data driving and physical modeling fused undercarriage buffer strut evaluation and prediction method, and the method comprises the steps: obtaining a corresponding attack angle value of an aircraft at a current grounding moment, and determining a corresponding initial lift coefficient of the aircraft at the current grounding moment based on the attack angle value; determining a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the attenuation duration of the lift coefficient; second objective functions corresponding to at least two acting force parameters of the aircraft in the vertical direction at the current grounding moment are determined, and delay time corresponding to undercarriage wheel load signals of the aircraft is determined according to the first objective functions and the second objective functions; and inputting the delay time into the prediction model for processing, obtaining target delay time corresponding to the undercarriage wheel load signal at the target moment, and determining a fault prediction result corresponding to the undercarriage buffer strut of the aircraft based on the target delay time.

Description

Technical Field

[0001] The embodiments of this specification relate to the technical field of fault prediction, and in particular to a landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling. Background Art

[0002] The delay of the landing gear wheel-load signal is the most important type of failure of the landing gear buffer strut, which is related to the safety of the aircraft. In actual flight, the aircraft may glide after landing until the front landing gear touches the ground, and the ground break lift function is still not turned on. The reason for this is that after the aircraft touches the ground, the main landing gear does not generate a wheel-load signal in time, and the ground break lift work requires the wheel-load signal, that is, the aircraft glides for a period of time after landing, and after the main landing gear wheel-load signal appears, the ground break lift function is immediately turned on and works normally.

[0003] During the flight of an aircraft, if the ground lift function is not turned on in time after the aircraft lands, it may be because the wheel-load signal is not generated in time after the landing main landing gear touches the ground. After the aircraft lands and taxis for a period of time, the ground lift function is turned on when the wheel-load signal appears. If the wheel-load signal is delayed, it will affect the ground lift function, reverse thrust, braking and other functions, resulting in an increase in the taxiing distance, which in turn affects flight safety. It can be seen that how to determine the delay time of the wheel-load signal and predict the fault of the aircraft's landing gear buffer strut based on the delay time has become an urgent problem to be solved. Summary of the invention

[0004] In view of this, the embodiments of this specification provide a landing gear buffer strut evaluation prediction method that integrates data drive and physical modeling. One or more embodiments of this specification also relate to a landing gear buffer strut evaluation prediction device that integrates data drive and physical modeling, a computing device, a computer readable storage medium, and a computer program to solve the technical defects existing in the prior art.

[0005] According to a first aspect of an embodiment of this specification, a landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling is provided, comprising: Acquire an angle of attack value corresponding to the current touchdown moment of the aircraft, and determine an initial lift coefficient corresponding to the current touchdown moment of the aircraft based on the angle of attack value; Determining a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the decay time of the lift coefficient; Determine second objective functions corresponding to at least two force parameters in the vertical direction of the aircraft at the current touchdown moment, and determine a delay time corresponding to a landing gear wheel load signal of the aircraft according to the first objective function and the second objective function; The delay time is input into a prediction model for processing to obtain a target delay time corresponding to the landing gear wheel load signal at a target time, and a fault prediction result corresponding to the landing gear buffer strut of the aircraft is determined based on the target delay time.

[0006] Optionally, the at least two force parameters include: load, landing gear buffer strut air cavity support force, landing gear buffer strut oil hole damping force; Accordingly, the method further comprises: Determine a second objective function corresponding to the air cavity support force of the landing gear buffer strut based on the initial inflation pressure of the landing gear buffer strut, the air cavity cross-sectional area and the landing gear buffer strut stroke; A second objective function corresponding to the oil hole damping force of the landing gear buffer strut is determined based on the oil hole damping coefficient and the compression speed of the landing gear buffer strut.

[0007] Optionally, determining the initial lift coefficient of the aircraft corresponding to the current touchdown moment based on the angle of attack value includes: Obtain the functional relationship between lift coefficient and angle of attack value; Based on the angle of attack value and the functional relationship, an initial lift coefficient corresponding to the aircraft at a current touchdown moment is determined.

[0008] Optionally, determining the delay time corresponding to the landing gear wheel-load signal of the aircraft according to the first objective function and the second objective function includes: Constructing a third objective function based on the first objective function and the second objective function; By solving the third objective function, the delay time corresponding to the landing gear wheel load signal of the aircraft is obtained.

[0009] Optionally, determining a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the attenuation time of the lift coefficient includes: The decay time of the lift coefficient is taken as the independent variable, the lift value of the aircraft at different times within the decay time is taken as the dependent variable, and based on the initial lift coefficient and the decay coefficient, a first objective function corresponding to the lift value of the aircraft is constructed.

[0010] Optionally, the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling further includes: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel-load signal, the parameter value interval is updated to generate a target parameter value interval.

[0011] Optionally, the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling further includes: Acquire multiple historical delay times corresponding to landing gear wheel load signals of the aircraft; Calculate the mean and standard deviation of the multiple historical delay times; Based on the mean, the standard deviation and the target confidence, a parameter value interval corresponding to the landing gear wheel load signal is constructed.

[0012] Optionally, the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling further includes: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel load signal, a health state corresponding to the delay time is constructed.

[0013] Optionally, the constructing the health status corresponding to the delay time includes: Acquire multiple historical delay times corresponding to the landing gear wheel load signal of the aircraft, and calculate the average value corresponding to the multiple historical delay times; Determining an absolute value of a difference between the delay time and the mean value; The absolute value is mapped to a target interval, and the mapping result is determined as a health state corresponding to the delay time, wherein an upper limit value and a lower limit value of the target interval are 1 and 0, respectively.

[0014] According to a second aspect of the embodiments of this specification, a landing gear buffer strut evaluation and prediction device integrating data-driven and physical modeling is provided, comprising: an acquisition module, configured to acquire an angle of attack value corresponding to the aircraft at a current touchdown moment, and determine an initial lift coefficient corresponding to the aircraft at the current touchdown moment based on the angle of attack value; A first determination module is configured to determine a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the decay time of the lift coefficient; A second determination module is configured to determine second objective functions corresponding to at least two force parameters in the vertical direction of the aircraft at a current touchdown moment, and determine a delay time corresponding to a landing gear wheel load signal of the aircraft according to the first objective function and the second objective function; The third determination module is configured to input the delay time into a prediction model for processing, obtain a target delay time corresponding to the landing gear wheel load signal at a target time, and determine a fault prediction result corresponding to the landing gear buffer strut of the aircraft based on the target delay time.

[0015] According to a third aspect of an embodiment of this specification, a computing device is provided, including: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement any one of the steps of the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling.

[0016] According to a fourth aspect of an embodiment of the present specification, there is provided a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of any one of the landing gear buffer strut evaluation and prediction methods integrating data-driven and physical modeling.

[0017] According to a fifth aspect of the embodiments of this specification, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling.

[0018] The data-driven and physical modeling-integrated landing gear buffer strut evaluation prediction method provided in the embodiment of this specification obtains the angle of attack value corresponding to the aircraft at the current touchdown moment, and determines the initial lift coefficient corresponding to the aircraft at the current touchdown moment based on the angle of attack value; determines the first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the attenuation time of the lift coefficient; determines the second objective function corresponding to at least two force parameters in the vertical direction of the aircraft at the current touchdown moment, and determines the delay time corresponding to the landing gear wheel-loaded signal of the aircraft according to the first objective function and the second objective function; inputs the delay time into the prediction model for processing, obtains the target delay time corresponding to the landing gear wheel-loaded signal at the target moment, and determines the fault prediction result corresponding to the landing gear buffer strut of the aircraft based on the target delay time. In this way, the delay time corresponding to the landing gear wheel-loaded signal of the aircraft is determined, and the fault prediction of the landing gear buffer strut is performed based on the delay time, and a fault warning can be performed when a fault is determined, which is conducive to avoiding the situation where the ground lift-breaking function cannot be opened in time due to the delay of the landing gear wheel-loaded signal, thereby helping to avoid or reduce the flight failure of the aircraft and improve flight safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of a landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling provided by an embodiment of this specification; Figure 2a It is a schematic diagram of the relationship between the radio altitude and the braking time corresponding to the braking phase of an aircraft provided by an embodiment of this specification; Figure 2b is a schematic diagram of a stable section of radio altitude provided by an embodiment of this specification; Figure 3 This is a schematic diagram of the relationship between the ground speed and the braking time corresponding to the braking phase of an aircraft provided by an embodiment of this specification; Figure 4 is a schematic diagram of the displacement of a landing gear buffer strut changing with time provided by an embodiment of the present specification; Figure 5 It is a schematic diagram of a prediction model provided by an embodiment of this specification; Figure 6 is a schematic diagram of a trend prediction result corresponding to the delay time of a wheel load signal provided by an embodiment of this specification; Figure 7a is a schematic diagram of a fitting result corresponding to the mean value of the historical delay time provided by an embodiment of this specification; Figure 7b is a schematic diagram of a fitting update result corresponding to the mean value of the historical delay time provided by an embodiment of this specification; Figure 7c is a schematic diagram of a fitting result corresponding to the standard deviation of the historical delay time provided in one embodiment of the present specification; Figure 7d is a schematic diagram of a fitting update result corresponding to the standard deviation of the historical delay time provided by an embodiment of this specification; Figure 8 It is a structural schematic diagram of a landing gear buffer strut evaluation and prediction device integrating data drive and physical modeling provided by an embodiment of this specification; Fig. 9 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0020] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0021] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0022] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0023] First, the terms involved in one or more embodiments of this specification are explained.

[0024] The embodiments of this specification consider the physical mechanism of the buffer strut compression process when the aircraft touches the ground, perform kinematic and dynamic modeling on the buffer strut compression process to obtain the landing gear wheel load signal delay time as a characteristic value for monitoring the health status of the landing gear buffer strut, and then combine Bayesian theory, deep recurrent networks and other data-driven algorithms to achieve health assessment and fault trend prediction of the landing gear buffer strut.

[0025] In this specification, a landing gear buffer strut evaluation and prediction method that integrates data-driven and physical modeling is provided. This specification also involves a landing gear buffer strut evaluation and prediction device that integrates data-driven and physical modeling, a computing device, a computer-readable storage medium, and a computer program, which are described in detail one by one in the following embodiments.

[0026] Figure 1 A flowchart of a landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling provided according to an embodiment of the present specification is shown, which specifically includes the following steps.

[0027] Step 102: Obtain an angle of attack value corresponding to the current touchdown moment of the aircraft, and determine an initial lift coefficient corresponding to the current touchdown moment of the aircraft based on the angle of attack value.

[0028] In an optional implementation, determining the initial lift coefficient corresponding to the aircraft at the current touchdown time based on the angle of attack value includes: Obtain the functional relationship between lift coefficient and angle of attack value; Based on the angle of attack value and the functional relationship, an initial lift coefficient corresponding to the aircraft at a current touchdown moment is determined.

[0029] Specifically, the current touchdown time is the instantaneous time corresponding to the moment when the aircraft touches down. For an aircraft, the angle of attack refers to the angle between the lift direction vector of the aircraft and the longitudinal axis of the aircraft.

[0030] The schematic diagram of the relationship between the radio altitude and the braking time corresponding to the braking phase of the aircraft provided in the embodiment of this specification is as follows: Figure 2a The schematic diagram of the stable section of the radio altitude provided in the embodiment of this specification is as shown in Figure 2b shown.

[0031] In the embodiments of this specification, the radio altitude signal 'RADIO_GEIGHT_L' can be used to capture the radio altitude stable segment in each flight segment using a parameter stable segment capture algorithm. Each flight segment contains multiple radio altitude stable segments, including a pre-takeoff stage, a cruise stage, and a post-landing stage. The last stable segment of each flight segment should be the stage where the radio altitude is 0 after the aircraft lands. Then, the timestamp of the aircraft's landing moment can be found, and this moment is recorded as the current touchdown moment of the aircraft.

[0032] In addition, the lift coefficient of the aircraft during the landing process There is a correlation between the angle of attack value of the aircraft during the landing process. When the angle of attack value is between -5 and 10 degrees, the two approximately satisfy a linear relationship, and in the embodiment of this specification, while considering the impact of the landing configuration and the ground effect on the lift coefficient when approaching the ground, the python data fitting function can be used to fit the result. The functional relationship with the angle of attack AOA is as follows: Formula (2) in, is the angle of attack value; is the correction value, K and are all constants, and K can be obtained by fitting.

[0033] In practical applications, multiple angle of attack values ​​and the lift coefficient corresponding to each angle of attack value can be obtained from the information contained in the flight manual of the aircraft. , and the correction value corresponding to each angle of attack value can be obtained , and then through each angle of attack value and each lift coefficient The corresponding relationship between them is obtained by fitting the value of K, that is, and The functional relationship between them is ; In addition, each angle of attack value and each correction value can be The corresponding relationship between them is obtained by fitting and The functional relationship between them is then summed to obtain the lift coefficient shown in formula (2): and angle of attack The functional relationship between .

[0034] When the functional relationship between the lift coefficient and the angle of attack value is determined, that is, the above formula (1), the angle of attack value corresponding to the current touchdown moment of the aircraft can be input into formula (1) to obtain the initial lift coefficient corresponding to the current touchdown moment of the aircraft. .

[0035] Step 104: Determine a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the decay time of the lift coefficient.

[0036] In an optional implementation, determining a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the attenuation time of the lift coefficient includes: The decay time of the lift coefficient is taken as the independent variable, the lift value of the aircraft at different times within the decay time is taken as the dependent variable, and based on the initial lift coefficient and the decay coefficient, a first objective function corresponding to the lift value of the aircraft is constructed.

[0037] Specifically, due to the lift coefficient The aircraft also changes in real time during the landing phase. Another way to calculate is: first find the angle of attack value of the aircraft at the moment of landing, and use the fitting algorithm to obtain the initial lift coefficient at the moment of landing , then the lift coefficient It decays quickly to 0, and the decay mode is linear decay. Taking the decay time t of the lift coefficient as the independent variable, we can get The expression is as follows: Formula (2) Wherein, k is the attenuation coefficient, T is the total attenuation time, and both k and T are constants.

[0038] The attenuation time of the lift coefficient is taken as the independent variable, and the lift coefficient As the dependent variable, determine the lift coefficient After the functional relationship between the initial lift coefficient and the decay time t is obtained, the decay time of the lift coefficient is taken as the independent variable, the lift value L of the aircraft at different times during the decay time is taken as the dependent variable, and based on the initial lift coefficient and the decay coefficient, the first objective function corresponding to the lift value of the aircraft is constructed as follows: Formula (3) in, is the air density; is the reference area of ​​the wing; v1 is the ground speed, and its value can be obtained by detecting the ground speed at different times through sensors; and are all constants.

[0039] The schematic diagram of the relationship between the ground speed and the braking time corresponding to the braking phase of the aircraft provided in the embodiment of this specification is as follows: Figure 3 shown.

[0040] In formula (2), Substituting the expression into formula (3), we can get the first objective function corresponding to the lift value of the aircraft.

[0041] In practical applications, the wing reference area You can take 79.86m 2 .

[0042] Step 106: Determine the second objective function corresponding to at least two force parameters in the vertical direction of the aircraft at the current touchdown moment, and determine the delay time corresponding to the landing gear wheel load signal of the aircraft based on the first objective function and the second objective function.

[0043] In an optional embodiment, the at least two force parameters include: load, landing gear buffer strut air cavity support force, landing gear buffer strut oil hole damping force; Accordingly, the method further comprises: Determine a second objective function corresponding to the air cavity support force of the landing gear buffer strut based on the initial inflation pressure of the landing gear buffer strut, the air cavity cross-sectional area and the landing gear buffer strut stroke; A second objective function corresponding to the oil hole damping force of the landing gear buffer strut is determined based on the oil hole damping coefficient and the compression speed of the landing gear buffer strut.

[0044] Further, determining the delay time corresponding to the landing gear wheel-load signal of the aircraft according to the first objective function and the second objective function includes: Constructing a third objective function based on the first objective function and the second objective function; By solving the third objective function, the delay time corresponding to the landing gear wheel load signal of the aircraft is obtained.

[0045] In the embodiments of this specification, the process of generating the landing gear wheel load signal when the aircraft touches down is actually a process of compression of the landing gear buffer strut, accompanied by the compression of the tire. The energy of the compression of the two comes from the impact kinetic energy of the aircraft sinking and touching down, as well as the work done by the combined force of lift and gravity in this process. Its characteristics are: 1) The energy of the aircraft sinking is absorbed by the landing gear buffer struts and tires, and the energy conversion is conserved; 2) When the landing gear buffer strut is compressed, part of the energy is stored in the compressed gas, and the other part is dissipated with the damping effect of oil and friction; 3) When a tire is compressed, it absorbs energy, and this energy is basically not dissipated.

[0046] Furthermore, considering the most severe situation of the landing gear wheel load signal, that is, considering the situation of the aircraft drifting down when touching the ground (sinking speed is 0), the aircraft part above the air cavity of the main landing gear buffer strut is taken as the research object, and its bouncing process, that is, the compression process of the landing gear buffer strut is analyzed.

[0047] In the embodiments of this specification, the vertical force acting on the aircraft when it touches the ground includes the lift L, and also includes the load G when the aircraft touches the ground, the air cavity support force f1 of the landing gear buffer strut, and the damping force f2 of the oil hole of the landing gear buffer strut.

[0048] Among them, the load G when the aircraft touches the ground is actually related to multiple factors such as the attitude and pitch angle of the aircraft when it touches the ground. The actual load calculation is relatively complicated. Therefore, in the embodiments of this specification, the load G is calculated according to the total gravity of the aircraft.

[0049] In addition, the second objective function corresponding to the air cavity support force of the landing gear buffer strut determined based on the initial inflation pressure of the landing gear buffer strut, the air cavity cross-sectional area and the landing gear buffer strut stroke is: Formula (4) in, Initial inflation pressure for landing gear cushion struts; is the cross-sectional area of ​​the air cavity; is the vertical displacement of the landing gear buffer strut; P and are all constants.

[0050] The schematic diagram of the displacement of the landing gear buffer strut changing with time provided in the embodiment of this specification is as follows Figure 4 shown.

[0051] In practical applications, the initial inflation pressure P can be set to 21 / 23 bar; the cross-sectional area of ​​the air cavity The value can be 0.017471749m 2 .

[0052] In addition, the second objective function corresponding to the oil hole damping force of the landing gear buffer strut determined based on the oil hole damping coefficient and the compression speed of the landing gear buffer strut is: Formula (5) Where, j is the damping coefficient of the oil hole; is the compression velocity of the buffer strut, and j is a constant.

[0053] In an embodiment of the present specification, after determining the first objective function or the second objective function corresponding to each force parameter in the vertical direction of the aircraft at the current touchdown moment, the delay time corresponding to the landing gear wheel load signal of the aircraft can be determined based on the first objective function and the second objective function.

[0054] Specifically, since the force parameters in the vertical direction of the aircraft satisfy a certain dynamic relationship, a third objective function can be constructed based on the first objective function and the second objective function, and by solving the third objective function, the delay time corresponding to the landing gear wheel load signal of the aircraft can be obtained.

[0055] Among them, the third objective function constructed based on the first objective function and the second objective function is: a Formula (6) Formula (6) includes the landing gear buffer strut stroke, that is, the displacement s of the landing gear buffer strut in the vertical direction, the buffer strut compression speed v2 and the acceleration a, and is a second-order differential equation. The embodiment of this specification can be solved by Python programming, and the specific algorithm is as follows: Step 1: Determine the initial values ​​and boundary conditions.

[0056] Initial value setting: s=0, v2=0, , L0 according to work out; Boundary conditions: threshold_s = 0.0254 (m) Step 2: Set the time step t=0.0001s.

[0057] Step 3: Every time a time step is advanced, the program updates the displacement s, real-time acceleration a, and velocity v2 until s > threshold_s, then stops the calculation and records the total time step, thus obtaining the delay time corresponding to the landing gear wheel load signal of the aircraft.

[0058] It should be noted that even if the landing gear buffer strut of the aircraft is compressed while the tire is compressed at the same time, the embodiment of this specification does not take into account the force generated by the tire compression when constructing the third objective function based on the dynamic relationship between the various force parameters in the vertical direction of the aircraft.

[0059] Step 108: Input the delay time into the prediction model for processing, obtain the target delay time corresponding to the landing gear wheel load signal at the target time, and determine the fault prediction result corresponding to the landing gear buffer strut of the aircraft based on the target delay time.

[0060] Specifically, after obtaining the delay time corresponding to the landing gear wheel load signal at the current landing moment of the aircraft, the delay time can be input into the pre-trained prediction model to predict the target delay time corresponding to the landing gear wheel load signal at the target moment through the prediction model, and based on the target delay time, the landing gear buffer strut of the aircraft is predicted to obtain the corresponding fault prediction result.

[0061] The prediction model of the embodiment of this specification adopts an end-to-end time series prediction model architecture, and uses a Bayesian neural network to quantify the prediction uncertainty. Specifically, a long short-term memory network (LSTM) is introduced to model the time series end-to-end, and the long-term dependency features existing in the time series are extracted. Uncertainty is introduced into the deep learning model from a Bayesian perspective, and random inactivation (dropouts) is applied after each hidden layer. The model output can be approximately regarded as a random sample generated by the posterior prediction distribution. Therefore, the uncertainty of the model can be estimated by the sample variance of several repeated model predictions.

[0062] A schematic diagram of a prediction model provided in the embodiments of this specification is as follows Figure 5 As shown, the model consists of two main components: (1) An encoder-decoder framework that captures the inherent patterns in the time series, which are learned during the pre-training process. The embodiments of this specification use an encoder-decoder framework with two layers of long short-term memory (LSTM) units.

[0063] (2) A prediction network that takes input from the learned embeddings of the encoder-decoder, as well as any potential external features to guide the prediction. In the embodiments of this specification, a multilayer perceptron is used as the prediction network.

[0064] The pre-training process of the prediction model is to first filter and smooth the multiple historical delay times of the landing gear wheel load signal, and then set the prediction time step of the model to 1 and the historical time step to 5. The prediction time step and the historical time step can be adjusted according to the characteristics of the actual data. Then, the constructed samples are first input into the encoder-decoder (AE) for pre-training, and then the features of the AE output are sent to the prediction network to obtain the trained prediction model.

[0065] After the complete model training process is completed, the prediction model obtained through training can be used to predict the trend of the delay time of the landing gear wheel load signal. Specifically, all historical delay times and the delay time corresponding to the landing gear wheel load signal at the current landing moment of the aircraft can be used to predict the target delay time at the next moment, that is, the target moment.

[0066] A schematic diagram of a trend prediction result corresponding to the delay time of a wheel load signal provided in an embodiment of this specification is as follows Figure 6 shown.

[0067] In addition, it is also possible to determine whether the landing gear buffer strut has a fault at the target time by determining whether the target delay time is within the parameter value interval corresponding to the preset landing gear wheel load signal; if it is determined that the target delay time is within the parameter value interval corresponding to the preset landing gear wheel load signal, it is determined that the landing gear buffer strut has no fault at the target time; otherwise, it is determined that the landing gear buffer strut has a fault at the target time. In this case, fault warning and other measures can be taken.

[0068] In practical applications, the real flight segment data of 46 flights can be used. The flight segment data is time series data with a sampling frequency of 1Hz and a sampling length of about 3600 to 10800 points. The flight parameters used include radio altitude (RADIO_GEIGHT), left inner wheel speed (WHL_SPD_BCU1(IN)_L), left outer wheel speed (WHL_SPD_BCU2(OUT)_L), right inner wheel speed (WHL_SPD_BCU1(IN)_R), right outer wheel speed (WHL_SPD_BCU2(OUT)_R), aircraft weight (GROSS_WEIGHT), ground speed (GRD_SPD). Other data include wing reference area, initial inflation pressure of buffer strut, air cavity cross-sectional area, oil hole damping coefficient, buffer strut compression speed parameters, etc.

[0069] In an optional embodiment, the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling further includes: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel-load signal, the parameter value interval is updated to generate a target parameter value interval.

[0070] Furthermore, the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling further includes: Acquire multiple historical delay times corresponding to landing gear wheel load signals of the aircraft; Calculate the mean and standard deviation of the multiple historical delay times; Based on the mean, the standard deviation and the target confidence, a parameter value interval corresponding to the landing gear wheel load signal is constructed.

[0071] Specifically, after determining the delay time corresponding to the landing gear wheel-loaded signal of the aircraft, it is also possible to first determine whether the delay time belongs to the parameter value interval corresponding to the pre-generated landing gear wheel-loaded signal; if so, step 108 can be executed, and the parameter value interval can be updated based on the delay time to generate a target parameter value interval. Specifically, when it is determined that the delay time belongs to the parameter value interval corresponding to the pre-generated landing gear wheel-loaded signal, the delay time is included in the historical delay time, and the upper and lower limit thresholds of the aforementioned parameter value interval are updated according to all historical delay times to adapt to the changing trend of the health status of the aircraft hydraulic system.

[0072] In the embodiment of this specification, since the delay time of the landing gear wheel load signal approximately obeys the normal distribution, the normal distribution is used to describe the delay time. The distribution parameters of the normal distribution are obtained by the Bayesian method. After the distribution parameters are obtained using the historical delay time, the upper and lower limit thresholds of the pre-constructed parameter value interval are updated based on the Bayesian theory using the newly launched flight segment data.

[0073] The specific process of constructing the upper and lower thresholds of the parameter value interval is as follows: 1) Firstly, a set of approximate distribution parameters are given for the mean and variance of the delay time of the landing gear wheel load signal based on experience. Then, based on the Bayesian method, the prior distribution of the mean and variance is adjusted using the historical delay time to make it approximate to the true distribution.

[0074] 2) Calculate the mean of the distribution of the mean and variance, and use the mean as the mean and variance of the delay time distribution of the landing gear wheel load signal. Select a reasonable confidence level, and use the confidence interval of the delay time of the landing gear wheel load signal as the upper and lower thresholds. The confidence interval is calculated as follows: Formula (7) Among them, up is the upper threshold, down is the lower threshold, is the average delay time, is the standard deviation of the delay time. For standard normal distribution Percentile, for Percentile.

[0075] In practical applications, the delay time of the landing gear wheel load signal can first be fitted based on the single normal distribution using Bayesian optimization. It follows a normal distribution, with a mean equal to the mean of the historical delay time of the historical flight segments and a variance of 1. No negative value, set the prior distribution The posterior distribution fitting result is obtained by fitting the above characteristic data according to the half-normal distribution with a standard deviation of 1. Further, based on the obtained posterior distribution, based on 3 The upper and lower thresholds of the parameter value interval are determined according to the principle.

[0076] After the upper and lower limit thresholds of the parameter value interval are constructed, the parameter value interval can be used to perform fault detection and health status judgment on the delay time of the landing gear wheel-loaded signal. When the newly-launched flight segment data enters, the delay time of the landing gear wheel-loaded signal is first extracted based on the above steps, and then it is determined whether the delay time is within the constructed upper and lower limit thresholds. If it is within the upper and lower limit thresholds, it is considered that the newly-launched flight segment data is in a normal state. If it exceeds the upper and lower limit thresholds, it is considered that the newly-launched flight segment data is in an abnormal state.

[0077] In addition, when the newly-launched segment data is in a normal state, it is necessary to include the delay time of the landing gear wheel load signal in the newly-launched segment data in the historical data, and use the extracted delay time of the landing gear wheel load signal to update the upper and lower thresholds. In order to avoid cumbersome calculations caused by too much historical data, the Bayesian theory is used to update the upper and lower thresholds and distribution parameters. Specifically, when the newly-launched segment data is judged to be in a normal state, the distribution parameters and the upper and lower thresholds are updated using the extracted delay time of the landing gear wheel load signal and a small part of the later historical delay time, rather than using all the historical delay times to update the upper and lower thresholds.

[0078] A schematic diagram of the fitting result corresponding to the mean value of the historical delay time provided in one embodiment of this specification is as follows: Figure 7a As shown; a schematic diagram of the fitting update result corresponding to the mean value of the historical delay time provided by an embodiment of this specification is as shown Figure 7b As shown, HDI represents the confidence level, mu_norm represents the normal distribution of the mean, and mean represents the mean; a schematic diagram of the fitting result corresponding to the standard deviation of the historical delay time provided in an embodiment of this specification is shown as Figure 7c As shown; a schematic diagram of the fitting update result corresponding to the standard deviation of the historical delay time provided by an embodiment of this specification is as shown Figure 7d As shown, sigma_norm represents the normal distribution of standard deviation.

[0079] In an optional embodiment, the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling further includes: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel load signal, a health state corresponding to the delay time is constructed.

[0080] Furthermore, the constructing of the health status corresponding to the delay time includes: Acquire multiple historical delay times corresponding to the landing gear wheel load signal of the aircraft, and calculate the average value corresponding to the multiple historical delay times; Determining an absolute value of a difference between the delay time and the mean value; The absolute value is mapped to a target interval, and the mapping result is determined as a health state corresponding to the delay time, wherein an upper limit value and a lower limit value of the target interval are 1 and 0, respectively.

[0081] Specifically, when it is determined that the delay time belongs to the parameter value interval corresponding to the pre-generated landing gear wheel load signal, a health state corresponding to the delay time can also be constructed, wherein the health state represents the degree of deviation of the delay time from the statistical mean.

[0082] In the embodiment of this specification, the health status construction process corresponding to the delay time is as follows: 1) First, get the absolute value of the difference between the delay time and its mean: Formula (8) Among them, e is the difference, x is the delay time, is the mean delay time.

[0083] 2) Since the difference between different delay times varies in size, the difference needs to be mapped to the interval (0, 1).

[0084] The mapping method can be a linear mapping: Formula (9) in, is the health degree, e is the difference, is the standard deviation of e. J is the health cutoff coefficient, which takes a value of (0,1), indicating that an early warning is required when the health is 1-J.

[0085] In practical applications, the health cutoff coefficient J can be 0.4, which means that an early warning is required when the health level is 0.6.

[0086] The mapping method can also be nonlinear: Formula (10) in, is the health degree, e is the difference, and J is the health degree cutoff coefficient, which is , Represents a health level of 1- Warning is needed when is the standard deviation of e.

[0087] The data-driven and physical modeling-integrated landing gear buffer strut evaluation prediction method provided in the embodiment of this specification obtains the angle of attack value corresponding to the aircraft at the current touchdown moment, and determines the initial lift coefficient corresponding to the aircraft at the current touchdown moment based on the angle of attack value; determines the first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the attenuation time of the lift coefficient; determines the second objective function corresponding to at least two force parameters in the vertical direction of the aircraft at the current touchdown moment, and determines the delay time corresponding to the landing gear wheel-loaded signal of the aircraft according to the first objective function and the second objective function; inputs the delay time into the prediction model for processing, obtains the target delay time corresponding to the landing gear wheel-loaded signal at the target moment, and determines the fault prediction result corresponding to the landing gear buffer strut of the aircraft based on the target delay time. In this way, the delay time corresponding to the landing gear wheel-loaded signal of the aircraft is determined, and the fault prediction of the landing gear buffer strut is performed based on the delay time, and a fault warning can be performed when a fault is determined, which is conducive to avoiding the situation where the ground lift-breaking function cannot be opened in time due to the delay of the landing gear wheel-loaded signal, thereby helping to avoid or reduce the flight failure of the aircraft and improve flight safety.

[0088] Corresponding to the above method embodiment, this specification also provides an embodiment of a landing gear buffer strut evaluation and prediction device integrating data-driven and physical modeling, Figure 8 The schematic diagram of the structure of a landing gear buffer strut evaluation and prediction device integrating data drive and physical modeling provided by an embodiment of the present specification is shown. Figure 8 As shown, the device comprises: The acquisition module 802 is configured to acquire an angle of attack value corresponding to the current touchdown moment of the aircraft, and determine an initial lift coefficient corresponding to the current touchdown moment of the aircraft based on the angle of attack value; A first determination module 804 is configured to determine a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the decay time of the lift coefficient; The second determination module 806 is configured to determine second objective functions corresponding to at least two force parameters in the vertical direction of the aircraft at the current touchdown moment, and determine a delay time corresponding to a landing gear wheel load signal of the aircraft according to the first objective function and the second objective function; The third determination module 808 is configured to input the delay time into the prediction model for processing, obtain the target delay time corresponding to the landing gear wheel load signal at the target time, and determine the fault prediction result corresponding to the landing gear buffer strut of the aircraft based on the target delay time.

[0089] Optionally, the at least two force parameters include: load, landing gear buffer strut air cavity support force, landing gear buffer strut oil hole damping force; Accordingly, the device further comprises a processing module configured to: Determine a second objective function corresponding to the air cavity support force of the landing gear buffer strut based on the initial inflation pressure of the landing gear buffer strut, the air cavity cross-sectional area and the landing gear buffer strut stroke; A second objective function corresponding to the oil hole damping force of the landing gear buffer strut is determined based on the oil hole damping coefficient and the compression speed of the landing gear buffer strut.

[0090] Optionally, the acquisition module 802 is further configured to: Obtain the functional relationship between lift coefficient and angle of attack value; Based on the angle of attack value and the functional relationship, an initial lift coefficient corresponding to the aircraft at a current touchdown moment is determined.

[0091] Optionally, the second determining module 806 is further configured to: Constructing a third objective function based on the first objective function and the second objective function; By solving the third objective function, the delay time corresponding to the landing gear wheel load signal of the aircraft is obtained.

[0092] Optionally, the first determining module 804 is further configured to: The decay time of the lift coefficient is taken as the independent variable, the lift value of the aircraft at different times within the decay time is taken as the dependent variable, and based on the initial lift coefficient and the decay coefficient, a first objective function corresponding to the lift value of the aircraft is constructed.

[0093] Optionally, the processing module is further configured to: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel-load signal, the parameter value interval is updated to generate a target parameter value interval.

[0094] Optionally, the processing module is further configured to: Acquire multiple historical delay times corresponding to landing gear wheel load signals of the aircraft; Calculate the mean and standard deviation of the multiple historical delay times; Based on the mean, the standard deviation and the target confidence, a parameter value interval corresponding to the landing gear wheel load signal is constructed.

[0095] Optionally, the processing module is further configured to: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel load signal, a health state corresponding to the delay time is constructed.

[0096] Optionally, the processing module is further configured to: Acquire multiple historical delay times corresponding to the landing gear wheel load signal of the aircraft, and calculate the average value corresponding to the multiple historical delay times; Determining an absolute value of a difference between the delay time and the mean value; The absolute value is mapped to a target interval, and the mapping result is determined as a health state corresponding to the delay time, wherein an upper limit value and a lower limit value of the target interval are 1 and 0, respectively.

[0097] The above is a schematic scheme of a landing gear buffer strut evaluation and prediction device that integrates data drive and physical modeling in this embodiment. It should be noted that the technical scheme of the landing gear buffer strut evaluation and prediction device that integrates data drive and physical modeling belongs to the same concept as the technical scheme of the landing gear buffer strut evaluation and prediction method that integrates data drive and physical modeling. For details that are not described in detail in the technical scheme of the landing gear buffer strut evaluation and prediction device that integrates data drive and physical modeling, please refer to the description of the technical scheme of the landing gear buffer strut evaluation and prediction method that integrates data drive and physical modeling.

[0098] Fig. 9 The structure block diagram of a computing device 900 provided according to an embodiment of the present specification is shown. The components of the computing device 900 include but are not limited to a memory 910 and a processor 920. The processor 920 is connected to the memory 910 via a bus 930, and a database 950 is used to store data.

[0099] The computing device 900 also includes an access device 940 that enables the computing device 900 to communicate via one or more networks 960. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 940 may include one or more of any type of network interface (e.g., a network interface card (NIC)) that is wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Wide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0100] In one embodiment of the present specification, the above components of the computing device 900 and Fig. 9Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Fig. 9 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0101] The computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. The computing device 900 may also be a mobile or stationary server.

[0102] The processor 920 is used to execute the following computer executable instructions, which, when executed by the processor, implement the steps of the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling.

[0103] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the landing gear buffer strut evaluation prediction method integrated with data drive and physical modeling belong to the same concept. For details not described in detail in the technical scheme of the computing device, please refer to the description of the technical scheme of the landing gear buffer strut evaluation prediction method integrated with data drive and physical modeling.

[0104] An embodiment of the present specification also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling.

[0105] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the landing gear buffer strut evaluation and prediction method integrated with data drive and physical modeling belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be found in the description of the technical scheme of the landing gear buffer strut evaluation and prediction method integrated with data drive and physical modeling.

[0106] An embodiment of the present specification also provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling.

[0107] The above is a schematic scheme of a computer program of this embodiment. It should be noted that the technical scheme of the computer program and the technical scheme of the landing gear buffer strut evaluation prediction method integrated with data drive and physical modeling belong to the same concept, and the details not described in detail in the technical scheme of the computer program can be found in the description of the technical scheme of the landing gear buffer strut evaluation prediction method integrated with data drive and physical modeling.

[0108] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0109] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0110] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0111] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0112] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. A landing gear buffer strut evaluation and prediction method based on data-driven and physical modeling integration, comprising: Acquire an angle of attack value corresponding to the current touchdown moment of the aircraft, and determine an initial lift coefficient corresponding to the current touchdown moment of the aircraft based on the angle of attack value; Determining a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the decay time of the lift coefficient; Determine second objective functions corresponding to at least two force parameters in the vertical direction of the aircraft at the current touchdown moment, and determine a delay time corresponding to a landing gear wheel load signal of the aircraft according to the first objective function and the second objective function; The delay time is input into a prediction model for processing to obtain a target delay time corresponding to the landing gear wheel load signal at a target time, and a fault prediction result corresponding to the landing gear buffer strut of the aircraft is determined based on the target delay time.

2. According to the landing gear buffer strut evaluation and prediction method based on data-driven and physical modeling fusion according to claim 1, the at least two force parameters include: Load, landing gear buffer strut air cavity support force, landing gear buffer strut oil hole damping force; Accordingly, the method further comprises: Determine a second objective function corresponding to the air cavity support force of the landing gear buffer strut based on the initial inflation pressure of the landing gear buffer strut, the air cavity cross-sectional area and the landing gear buffer strut stroke; A second objective function corresponding to the oil hole damping force of the landing gear buffer strut is determined based on the oil hole damping coefficient and the compression speed of the landing gear buffer strut.

3. According to the data-driven and physical modeling integrated landing gear buffer strut evaluation and prediction method of claim 1, the determining of the initial lift coefficient corresponding to the aircraft at the current touchdown moment based on the angle of attack value comprises: Obtain the functional relationship between lift coefficient and angle of attack value; Based on the angle of attack value and the functional relationship, an initial lift coefficient corresponding to the aircraft at a current touchdown moment is determined.

4. The landing gear buffer strut evaluation and prediction method based on data-driven and physical modeling integration according to claim 1, wherein determining the delay time corresponding to the landing gear wheel load signal of the aircraft according to the first objective function and the second objective function comprises: Constructing a third objective function based on the first objective function and the second objective function; By solving the third objective function, the delay time corresponding to the landing gear wheel load signal of the aircraft is obtained.

5. According to the data-driven and physical modeling integrated landing gear buffer strut evaluation and prediction method according to any one of claims 1 to 4, the first objective function corresponding to the lift value of the aircraft is determined based on the initial lift coefficient and the attenuation time of the lift coefficient, comprising: The decay time of the lift coefficient is taken as the independent variable, the lift value of the aircraft at different times within the decay time is taken as the dependent variable, and based on the initial lift coefficient and the decay coefficient, a first objective function corresponding to the lift value of the aircraft is constructed.

6. The landing gear buffer strut evaluation and prediction method based on data-driven and physical modeling integration according to claim 1, further comprising: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel-load signal, the parameter value interval is updated to generate a target parameter value interval.

7. The landing gear buffer strut evaluation and prediction method based on data-driven and physical modeling integration according to claim 1 or 6, further comprising: Acquire multiple historical delay times corresponding to landing gear wheel load signals of the aircraft; Calculate the mean and standard deviation of the multiple historical delay times; Based on the mean, the standard deviation and the target confidence, a parameter value interval corresponding to the landing gear wheel load signal is constructed.

8. The landing gear buffer strut evaluation and prediction method based on data-driven and physical modeling integration according to claim 1, further comprising: When it is determined that the delay time belongs to the pre-generated parameter value interval corresponding to the landing gear wheel load signal, a health state corresponding to the delay time is constructed.

9. According to the landing gear buffer strut evaluation and prediction method integrating data-driven and physical modeling as claimed in claim 8, the step of constructing the health state corresponding to the delay time comprises: Acquire multiple historical delay times corresponding to the landing gear wheel load signal of the aircraft, and calculate the average value corresponding to the multiple historical delay times; Determining an absolute value of a difference between the delay time and the mean value; The absolute value is mapped to a target interval, and the mapping result is determined as a health state corresponding to the delay time, wherein an upper limit value and a lower limit value of the target interval are 1 and 0, respectively.

10. A landing gear buffer strut evaluation and prediction device integrating data-driven and physical modeling, comprising: an acquisition module, configured to acquire an angle of attack value corresponding to the aircraft at a current touchdown moment, and determine an initial lift coefficient corresponding to the aircraft at the current touchdown moment based on the angle of attack value; A first determination module is configured to determine a first objective function corresponding to the lift value of the aircraft based on the initial lift coefficient and the decay time of the lift coefficient; A second determination module is configured to determine second objective functions corresponding to at least two force parameters in the vertical direction of the aircraft at a current touchdown moment, and determine a delay time corresponding to a landing gear wheel load signal of the aircraft according to the first objective function and the second objective function; The third determination module is configured to input the delay time into a prediction model for processing, obtain a target delay time corresponding to the landing gear wheel load signal at a target time, and determine a fault prediction result corresponding to the landing gear buffer strut of the aircraft based on the target delay time.

Images