A method for early warning of aircraft brake wheel temperature based on data processing
Through the variational mode decomposition algorithm and Attention-LSTM model, the temperature prediction and high temperature warning of aircraft brake wheels are performed, and the cooling device is dynamically controlled, which solves the problems of shortening the life of the brake device and high power consumption caused by untimely cooling in the prior art. It realizes automated and intelligent cooling of temperature warning of aircraft brake wheels, enhancing flight safety.
Patent Information
- Application Number
- CN202411273815.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-12
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-12
AI Technical Summary
The existing aircraft brake wheel temperature control scheme shortens the service life of the brake device when the cooling device is not cooled in time, and the cooling device has a high power consumption, which has energy saving problems.
By using a data processing method, temperature prediction and prediction are carried out by obtaining the temperature historical timing data of the aircraft brake wheel, the historical data of the empty state, the landing taxi state and the landing resting state of the ground, the variational modal decomposition algorithm and the Attention-LSTM model are used to predict the temperature and predict the temperature and High temperature warning, dynamically control the brake cooling device to achieve timely cooling.
It improves the accuracy and robustness of temperature prediction and high temperature warning, realizes the automation of temperature warning of aircraft brake wheels and the intelligence of cooling and cooling, saves energy and extends the service life of the brake device.
Smart Images

Figure CN118770168B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of aircraft brake wheel temperature control, and in particular relates to an aircraft brake wheel temperature early warning method based on data processing. Background Art
[0002] Aircraft brake wheels generally include a wheel assembly and a brake device, wherein the wheel assembly and the matching tire are used for the aircraft to park and taxi on the ground, and to withstand the impact load during landing, while the brake device provides braking torque when the aircraft brakes and absorbs the aircraft's horizontal kinetic energy, so that the aforementioned wheel assembly and brake device can provide the aircraft with dynamic braking and static braking functions.
[0003] Modern aircraft have the characteristics of heavy landing weight, large braking energy and high braking speed, so the existing brake devices are mainly made of C / C composite materials (because C / C composite materials have the advantages of good friction characteristics, strong heat absorption, low density and long service life), but C / C composite materials also have the defect of poor oxidation resistance, which makes the brake device unable to be water-cooled. At the same time, in order to ensure the rapid deployment of the aircraft, the brake device based on C / C composite materials and the entire aircraft brake wheel must be forced to cool, making the brake cooling device also a necessary accessory for the aircraft brake wheel.
[0004] At present, existing brake cooling devices are mostly implemented by heat dissipation fans and / or semiconductor refrigeration sheets, and their working principles require the consumption of a certain amount of electrical energy. Therefore, for the purpose of energy saving, existing brake cooling devices need to be activated only when certain high temperature conditions are reached. For example, the brake cooling device will be activated only when the temperature of the aircraft brake wheel reaches a preset temperature threshold to cool down the temperature. However, this temperature control method has an obvious lag disadvantage, which can easily shorten the service life of the brake device due to untimely cooling.
[0005] In summary, how to accurately predict the changing trend of aircraft brake wheel temperature and issue high temperature warnings so as to dynamically control the brake cooling device in a timely manner according to the warning conditions, thereby achieving the purpose of energy saving without affecting the service life of the brake device, is a topic that technical personnel in this field urgently need to study. Summary of the invention
[0006] The purpose of the present invention is to provide an aircraft brake wheel temperature warning method based on data processing, a computer device, a computer readable storage medium and a computer program product, so as to solve the problem that the existing aircraft brake wheel temperature control scheme is prone to shortening the service life of the brake device due to untimely cooling.
[0007] In order to achieve the above object, the present invention adopts the following technical solutions:
[0008] In a first aspect, a method for early warning of aircraft brake wheel temperature based on data processing is provided, comprising:
[0009] Obtain the temperature history time series data of the aircraft brake wheel, the history time series data of the empty state, the history time series data of the landing taxiing state and the history time series data of the landing static state, wherein the temperature history time series data contains the unit time period and in the the historical temperature value of each unit time period in each unit time period, the empty state historical time series data contains the binary empty state value in each unit time period, the binary empty state value is "1" indicating that the aircraft brake wheel is in the empty state in the corresponding unit time period, and the binary empty state value is "0" indicating that the aircraft brake wheel is not in the empty state in the corresponding unit time period, the landing taxiing state historical time series data contains the binary landing taxiing state value in each unit time period, the binary landing taxiing state value is "1" indicating that the aircraft brake wheel is in the landing taxiing state in the corresponding unit time period, and the binary landing taxiing state value is "0" indicating that the aircraft brake wheel is not in the landing taxiing state in the corresponding unit time period, the landing static state historical time series data contains the binary landing static state value in each unit time period, the binary landing static state value is "1" indicating that the aircraft brake wheel is in the landing static state in the corresponding unit time period, and the binary landing static state value is "0" indicating that the aircraft brake wheel is not in the landing static state in the corresponding unit time period, represents a positive integer greater than or equal to 3;
[0010] According to the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the aircraft brake wheel in the future is predicted. Estimated temperature value for unit time period, where Indicates less than A positive integer of ;
[0011] According to the aircraft brake wheel in the future The estimated temperature value of the aircraft brake wheel in the future The temperature warning level for each unit period.
[0012] Based on the above invention content, a new solution for aircraft brake wheel temperature prediction and high temperature warning based on temperature and state historical data is provided, that is, firstly, the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing stationary state historical time series data of the aircraft brake wheel are obtained, and then according to these data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the estimated temperature value of the aircraft brake wheel in the future period is predicted, and finally according to these estimated temperature values, the temperature warning level of the aircraft brake wheel in the future period is determined. In this way, on the one hand, by adopting a combination of multiple methods for prediction, the advantages of various methods can be fully utilized to improve the accuracy and robustness of the temperature value and the high temperature warning result. On the other hand, it can also be beneficial to timely and dynamically control the brake cooling device according to the warning situation, so as to achieve the purpose of energy saving without affecting the service life of the brake device, thereby realizing the automation of the aircraft brake wheel temperature warning process and the intelligence of cooling and cooling, enhancing flight safety, and facilitating practical application and promotion.
[0013] In a possible design, based on the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, the aircraft brake wheel in the future is predicted based on the variational mode decomposition algorithm and the Attention-LSTM model. Estimated temperature values for unit periods, including:
[0014] Decomposing the temperature history time series data by using a variational mode decomposition algorithm to obtain three mode component time series data, and making the three mode component time series data correspond to the empty state, the landing gliding state and the landing static state respectively;
[0015] For each of the three modal component time series data, a plurality of corresponding sample data are extracted from the corresponding time series data in the following manner: sequentially continuous sample data are extracted from the corresponding component time series data. The historical temperature value modal components of the unit time period and the corresponding state historical time series data are extracted in the unit time periods and the After unit periods and sequentially in time The values of the unit time periods are taken together as the model input items, and the corresponding component time series data are extracted in the The historical temperature value modal component of a unit period is used as the model output item, and then the model input item and the model output item are used as a sample data, wherein, Indicates greater than or equal to 2 and less than A positive integer, Indicates less than or equal to A positive integer of ;
[0016] For each of the modal component time series data, extract a plurality of training sample data from the corresponding plurality of sample data, and use the plurality of training sample data to perform model training on the Attention-LSTM model to obtain a corresponding temperature value prediction model;
[0017] According to the future flight plan of the aircraft to which the aircraft brake wheel belongs, the time series data of the aircraft brake wheel in the air, the time series data of the landing taxiing state and the time series data of the landing stationary state are derived, wherein the time series data of the aircraft brake wheel in the air contains the time series data which are sequentially continuous in time series. future unit periods and in the The binarization of each future unit time period in the future unit time period is in the empty state value, and the landing and taxiing state planning time series data contains the future unit periods and in the The binary landing and sliding state value of each future unit time period in the future unit time periods, the landing stationary state planning time series data contains the future unit periods and in the The binary ground stationary state value of each future unit time period in the future unit time periods, Indicates greater than or equal to A positive integer of ;
[0018] For each modal component time series data, extract the most recent consecutive time series data from the corresponding component time series data. The historical temperature value modal components of the unit period and the corresponding state historical time series data are extracted in the most recent The value of the unit time period and the nearest The future that is consecutive in time after unit periods The values of the unit time periods are imported into the corresponding temperature value prediction model together, and the corresponding model output prediction new item is output;
[0019] The model output prediction new item of each modal component time series data is superimposed to reconstruct the aircraft brake wheel in the future Estimated temperature value for a unit period.
[0020] In one possible design, the method further includes:
[0021] Acquire historical time series data of the heat dissipation working state of a brake cooling device used to cool the aircraft brake wheel, wherein the historical time series data of the heat dissipation working state contains working state parameter values of the brake cooling device in each unit time period;
[0022] According to the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the aircraft brake wheel in the future is predicted. Estimated temperature values for unit periods, including:
[0023] Decomposing the temperature history time series data by using a variational modal decomposition algorithm to obtain four modal component time series data, and making the four modal component time series data correspond to the empty state, the landing and gliding state, the landing and static state, and the heat dissipation working state respectively;
[0024] For each of the four modal component time series data, a plurality of corresponding sample data are extracted from the corresponding time series data in the following manner: the sample data that are sequentially continuous in time series are extracted from the corresponding component time series data. The historical temperature value modal components of the unit time period and the corresponding state historical time series data are extracted in the unit time periods and the After unit periods and sequentially in time The values of the unit time periods are taken together as the model input items, and the corresponding component time series data are extracted in the The historical temperature value modal component of a unit period is used as the model output item, and then the model input item and the model output item are used as a sample data, wherein, Indicates greater than or equal to 2 and less than A positive integer, Indicates less than or equal to A positive integer of ;
[0025] For each of the modal component time series data, extract a plurality of training sample data from the corresponding plurality of sample data, and use the plurality of training sample data to perform model training on the Attention-LSTM model to obtain a corresponding temperature value prediction model;
[0026] According to the future flight plan of the aircraft to which the aircraft brake wheel belongs, the planned time series data of the aircraft brake wheel in the air, the planned time series data of the landing taxiing state and the planned time series data of the landing stationary state are derived, and according to a working state parameter value of the most recent unit time period in the historical time series data of the heat dissipation working state, the planned time series data of the heat dissipation working state is derived, wherein the planned time series data of the air state contains the following data which are consecutive in time sequence: future unit periods and in the The binarization of each future unit time period in the future unit time period is in the empty state value, and the landing and taxiing state planning time series data contains the future unit periods and in the The binary landing and sliding state value of each future unit time period in the future unit time periods, the landing stationary state planning time series data contains the future unit periods and in the The binary landing static state value of each future unit time period in the future unit time period, the heat dissipation working state planning time series data contains the future unit periods and in the each future unit time period in the future unit time period and equal to the working state parameter value of the certain working state parameter value, Indicates greater than or equal to A positive integer of ;
[0027] For each modal component time series data, extract the most recent consecutive time series data from the corresponding component time series data. The historical temperature value modal components of the unit period and the corresponding state historical time series data are extracted in the most recent The value of the unit time period and the nearest The future that is consecutive in time after unit periods The values of the unit time periods are imported into the corresponding temperature value prediction model together, and the corresponding model output prediction new item is output;
[0028] The model output prediction new item of each modal component time series data is superimposed to reconstruct the aircraft brake wheel in the future Estimated temperature value for a unit period.
[0029] In a possible design, based on the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, the aircraft brake wheel in the future is predicted based on the variational mode decomposition algorithm and the Attention-LSTM model. Estimated temperature values for each unit period, including:
[0030] For each of the modal component time series data, a plurality of test sample data are extracted from the corresponding plurality of sample data, and a plurality of model input items in the plurality of test sample data are introduced into the corresponding temperature value prediction model, and a plurality of corresponding model output prediction items are output, wherein the model input item of each of the plurality of test sample data has the same value for all the modal component time series data and is sequentially continuous in time series. unit time period, the plurality of model input items correspond one-to-one to the plurality of sample data, and the plurality of model output prediction items correspond one-to-one to the plurality of model input items;
[0031] The same unit time periods, superimpose the time series data of each modal component and the corresponding model output prediction item, and reconstruct the time series data that is located after the corresponding time period and is sequentially continuous in time series. The estimated temperature value of each unit period, and the values that are located after the corresponding period and are consecutive in time sequence The historical temperature values of each unit period are subtracted one by one from the historical temperature values of the corresponding period and consecutive in time sequence. The estimated temperature value of each unit period is obtained after the corresponding period and is sequentially continuous in time sequence. Temperature error value per unit time period;
[0032] According to the same After unit periods and sequentially in time The temperature error value of each unit time period is constructed to obtain the temperature error time series data, wherein the temperature error time series data contains the temperature error values that are consecutive in time series. unit time period and in the The temperature error value of each unit time period in the unit time period is represents a positive integer greater than or equal to 3;
[0033] A plurality of new sample data are extracted from the temperature error time series data in the following manner: extracting the temperature error time series data that are sequentially continuous in time series. The temperature error value of each unit period is used as a new model input item, and the temperature error time series data is extracted from the After unit periods and sequentially in time The temperature error value of each unit period is used as a new model output item, and then the model input new item and the model output new item are used as a new sample data, wherein, , ;
[0034] Using the multiple new sample data to perform model training on the LSTM model to obtain a temperature error prediction model;
[0035] Extract the most recent The temperature error value of each unit period is imported into the temperature error prediction model to obtain the future Temperature error prediction value for each unit period;
[0036] Make the future The estimated temperature values of each unit period are added to the future The temperature error prediction value of the aircraft brake wheel in the future The final estimated temperature value for the unit period.
[0037] In one possible design, the aircraft brake wheel is The estimated temperature value of the aircraft brake wheel in the future The temperature warning level for each unit period includes:
[0038] According to the aircraft brake wheel in the future The estimated temperature values of the unit time periods are statistically obtained to obtain the maximum estimated temperature or the average estimated temperature;
[0039] According to a plurality of temperature value ranges that are preset and correspond one to one with a plurality of temperature warning levels, determining a temperature value range that includes the estimated maximum temperature value or the estimated average temperature value from the plurality of temperature value ranges;
[0040] A temperature warning level corresponding to the temperature value range is used as the aircraft brake wheel in the future The temperature warning level for each unit period.
[0041] In one possible design, the aircraft brake wheel is determined in the future After the temperature warning level of a unit period is determined, the method further comprises:
[0042] A preset working instruction corresponding to the temperature warning level is transmitted to the brake cooling device, so as to cool down the aircraft brake wheel by controlling the working state of the brake cooling device.
[0043] In a possible design, the brake cooling device includes a heat dissipation fan and / or a semiconductor cooling fin.
[0044] In a possible design, obtaining the temperature history time series data of the aircraft brake wheel includes:
[0045] Acquiring raw temperature data periodically collected by a temperature sensor installed in the aircraft brake wheel;
[0046] The original temperature data is preprocessed to obtain the temperature history time series data of the aircraft brake wheel, wherein the preprocessing includes data cleaning processing, data interpolation processing, data abnormality processing and / or data normalization processing, and the temperature history time series data contains the temperature history data which are sequentially continuous in time series. unit time period and in the The historical temperature values of each unit period in the unit period, Represents a positive integer greater than or equal to 3.
[0047] In a possible design, the acquisition of the historical time series data of the aircraft brake wheel in the air state, the historical time series data of the landing taxiing state, and the historical time series data of the landing stationary state includes:
[0048] According to the flight history record of the aircraft to which the aircraft brake wheel belongs, the historical time series data of the aircraft brake wheel in the air, the historical time series data of the landing taxiing state and the historical time series data of the landing static state are derived, wherein the historical time series data of the aircraft brake wheel in the air contains the following data: unit time period and in the The binary empty state value of each unit time period in the unit time period is "1", indicating that the aircraft brake wheel is in the empty state in the corresponding unit time period, and the binary empty state value is "0", indicating that the aircraft brake wheel is not in the empty state in the corresponding unit time period. The landing taxiing state historical time series data contains the binary landing taxiing state value in each unit time period, the binary landing taxiing state value is "1", indicating that the aircraft brake wheel is in the landing taxiing state in the corresponding unit time period, and the binary landing taxiing state value is "0", indicating that the aircraft brake wheel is in the non-landing taxiing state in the corresponding unit time period. The landing static state historical time series data contains the binary landing static state value in each unit time period, the binary landing static state value is "1", indicating that the aircraft brake wheel is in the landing static state in the corresponding unit time period, and the binary landing static state value is "0", indicating that the aircraft brake wheel is in the non-landing static state in the corresponding unit time period. Represents a positive integer greater than or equal to 3.
[0049] In a possible design, the parameters of the variational mode decomposition algorithm are tuned using a particle swarm algorithm, and the parameters of the Attention-LSTM model are searched using a grid search algorithm or a Bayesian search algorithm.
[0050] In a second aspect, the present invention provides a computer device comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the aircraft brake wheel temperature warning method based on data processing as described in the first aspect or any possible design of the first aspect.
[0051] In a third aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, executes the aircraft brake wheel temperature warning method based on data processing as described in the first aspect or any possible design of the first aspect.
[0052] In a fourth aspect, the present invention provides a computer program product, comprising a computer program or instructions, which, when executed by a computer, implements the aircraft brake wheel temperature warning method based on data processing as described in the first aspect or any possible design of the first aspect.
[0053] Beneficial effects of the above scheme:
[0054] (1) The present invention creatively provides a new solution for aircraft brake wheel temperature prediction and high temperature warning based on temperature and state historical data, that is, firstly obtaining the aircraft brake wheel temperature historical time series data, the air state historical time series data, the landing taxi state historical time series data and the landing stationary state historical time series data, and then predicting the estimated temperature value of the aircraft brake wheel in the future period based on these data, based on the variational mode decomposition algorithm and the Attention-LSTM model, and finally determining the temperature warning level of the aircraft brake wheel in the future period based on these estimated temperature values. In this way, on the one hand, by adopting a combination of multiple methods for prediction, the advantages of various methods can be fully utilized to improve the accuracy and robustness of the temperature value and high temperature warning results. On the other hand, it can also facilitate the timely dynamic control of the brake cooling device according to the warning situation, so as to achieve the purpose of energy saving without affecting the service life of the brake device, thereby realizing the automation of the aircraft brake wheel temperature warning process and the intelligence of cooling, enhancing flight safety, and facilitating practical application and promotion;
[0055] (2) By considering the heat dissipation working state of the brake cooling device, a more realistic and accurate aircraft brake wheel temperature prediction result can be obtained based on the variational mode decomposition algorithm and the Attention-LSTM model, further improving the accuracy and robustness of the warning results.
[0056] (3) The accuracy and robustness of the warning results can be further improved by constructing temperature error time series data, training a temperature error prediction model based on this data, and applying the temperature error prediction model to correct the estimated temperature values of the aircraft brake wheels in the future. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0058] Figure 1 A schematic flow chart of an aircraft brake wheel temperature warning method based on data processing provided in an embodiment of the present application.
[0059] Figure 2 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structures of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these embodiments without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.
[0061] It should be understood that although the terms first and second, etc. may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another object. For example, a first object can be referred to as a second object, and similarly, a second object can be referred to as a first object without departing from the scope of the exemplary embodiments of the present invention.
[0062] It should be understood that the term "and / or" that may appear in this document is merely a description of the association relationship between associated objects, indicating that there may be three relationships. For example, A and / or B can indicate three situations: A exists alone, B exists alone, or A and B exist at the same time. For another example, A, B and / or C can indicate the existence of any one of A, B and C or any combination of them. The term " / and" that may appear in this document describes another type of association object relationship, indicating that there may be two relationships. For example, A / and B can indicate two situations: A exists alone or A and B exist at the same time. In addition, the character " / " that may appear in this document generally indicates that the previous and next associated objects are in an "or" relationship.
[0063] Example
[0064] like Figure 1 As shown, the aircraft brake wheel temperature warning method provided in the first aspect of this embodiment and based on data processing can be executed by, but not limited to, a computer device with certain computing resources, such as an airborne server or a temperature controller of an aircraft brake wheel. Figure 1 As shown, the aircraft brake wheel temperature early warning method may include but is not limited to the following steps S1 to S3.
[0065] S1. Obtain the historical time series data of the temperature of the aircraft brake wheel, the historical time series data of the empty state, the historical time series data of the landing taxiing state and the historical time series data of the landing static state, wherein the temperature historical time series data includes but is not limited to the temperature historical time series data which are sequentially continuous in time series. unit time period and in the The historical temperature values of each unit time period in each unit time period, etc., the empty state historical time series data includes but is not limited to the binary empty state values in each unit time period, etc., the binary empty state value is "1" indicating that the aircraft brake wheel is in the empty state in the corresponding unit time period, the binary empty state value is "0" indicating that the aircraft brake wheel is not in the empty state in the corresponding unit time period, the landing and taxiing state historical time series data includes but is not limited to the binary landing and taxiing state values in each unit time period, etc., the binary landing and taxiing state value is "1" Indicates that the aircraft brake wheel is in the landing and taxiing state in the corresponding unit time period, the binary landing and taxiing state value is "0" indicating that the aircraft brake wheel is in the non-landing and taxiing state in the corresponding unit time period, the landing stationary state historical time series data includes but is not limited to the binary landing stationary state value in each unit time period, the binary landing stationary state value is "1" indicating that the aircraft brake wheel is in the landing stationary state in the corresponding unit time period, the binary landing stationary state value is "0" indicating that the aircraft brake wheel is in the non-landing stationary state in the corresponding unit time period, Represents a positive integer greater than or equal to 3.
[0066] In step S1, the unit time period may be, but is not limited to, 10 seconds, 30 seconds, or 1 minute, for example, that is, the temperature history time series data may include, but is not limited to, the temperature of the aircraft brake wheel in the most recent The historical temperature value of the minute, The historical temperature value of the last minute, ..., the historical temperature value of the last second minute and the historical temperature value of the last minute, etc.; the historical time series data in the empty state contains the historical temperature value of the aircraft brake wheel in the last minute. The binarization of the minute is in the empty state value, in the most recent The binarization of the last minute is in the empty state value, ..., the binarization of the last second minute is in the empty state value and the binarization of the last minute is in the empty state value, etc.; the landing taxiing state historical time series data contains the aircraft brake wheel in the most recent The binary landing and sliding state value of the minute, The binary landing taxiing state value in the last minute, ..., the binary landing taxiing state value in the last second minute and the binary landing taxiing state value in the last minute, etc.; the landing stationary state historical time series data contains the aircraft brake wheel in the last minute. The binary value of the landing static state of the minute, in the most recent The binary landing static state value of the last minute, ..., the binary landing static state value of the last second minute and the binary landing static state value of the last minute, etc. Specifically, obtaining the temperature history time series data of the aircraft brake wheel includes but is not limited to the following steps S11-S12.
[0067] S11. Acquire raw temperature data periodically collected by a temperature sensor installed in the aircraft brake wheel.
[0068] In step S11, the temperature sensor may be an existing product and conventionally installed in the aircraft brake wheel (for example, installed in a temperature sensor installation hole on the wheel assembly hub), and its acquisition period needs to be less than the duration of the unit time period, for example, 100 milliseconds.
[0069] S12. Preprocess the original temperature data to obtain the temperature history time series data of the aircraft brake wheel, wherein the preprocessing includes but is not limited to data cleaning processing, data interpolation processing, data abnormality processing and / or data normalization processing, and the temperature history time series data includes but is not limited to the temperature history data that are sequentially continuous in time series. unit time period and in the The historical temperature values of each unit time period in each unit time period, etc. Represents a positive integer greater than or equal to 3.
[0070] In step S12, the aforementioned historical temperature value may be, but is not limited to, the maximum temperature, the middle temperature, the median temperature or the final temperature value collected within the corresponding unit time period. The data cleaning process is used to clean the dirty data in the original temperature data, and can be implemented by existing conventional data cleaning means. The data interpolation process is used to fill in the missing data in the original temperature data, and can be implemented by existing conventional data interpolation means. The data anomaly processing is used to correct the abnormal data in the original temperature data, and can be implemented by existing conventional data correction means. The data normalization process is used to make the characteristic scale of the temperature historical time series data consistent, so as to improve the performance of the subsequent prediction model and the convergence speed of the model training, and can be implemented by existing conventional data normalization means.
[0071] In the step S1, since the aircraft brake wheel is in only one state in the same unit time period, the binary empty state value, the binary landing taxiing state value and the binary landing stationary state value need to ensure that any two state values are not "1" at the same time. Specifically, obtaining the empty state historical time series data, the landing taxiing state historical time series data and the landing stationary state historical time series data of the aircraft brake wheel includes but is not limited to the following step S13.
[0072] S13. According to the flight history record of the aircraft to which the aircraft brake wheel belongs, derive the historical time series data of the aircraft brake wheel in the air, the historical time series data of the landing and taxiing state, and the historical time series data of the landing and stationary state, wherein the historical time series data of the aircraft brake wheel in the air includes but is not limited to the historical time series data of the aircraft brake wheel in the air. unit time period and in the The binary empty state value of each unit time period in the unit time period is "1", indicating that the aircraft brake wheel is in an empty state (i.e., in the air) in the corresponding unit time period, and the binary empty state value is "0", indicating that the aircraft brake wheel is in a non-empty state (i.e., it may be a landing taxiing state or a landing stationary state) in the corresponding unit time period. The landing taxiing state historical time series data includes but is not limited to the binary landing taxiing state value in each unit time period, and the binary landing taxiing state value is "1", indicating that the aircraft brake wheel is in a landing taxiing state in the corresponding unit time period. The binary landing and taxiing state value is "0" which indicates that the aircraft brake wheel is in a non-landing and taxiing state (i.e., it may be in an empty state or a landing and stationary state) in the corresponding unit time period. The landing and stationary state historical time series data includes but is not limited to the binary landing and stationary state values in each unit time period. The binary landing and stationary state value is "1" which indicates that the aircraft brake wheel is in a landing and stationary state in the corresponding unit time period. The binary landing and stationary state value is "0" which indicates that the aircraft brake wheel is in a non-landing and stationary state (i.e., it may be in an empty state or a landing and taxiing state) in the corresponding unit time period. Represents a positive integer greater than or equal to 3.
[0073] In the step S13, since the flight history record contains the state of the aircraft to which the aircraft brake wheel belongs in each historical unit time period, the binary empty state value, the binary landing taxiing state value and the binary landing stationary state value in each historical unit time period can be conventionally derived based on the state. For example, when the aircraft to which the aircraft brake wheel belongs is in a flying state in the first historical unit time period, it can be derived that the binary empty state value in the first historical unit time period is "1", the binary landing taxiing state value in the first historical unit time period is "0", and the binary landing stationary state value in the first historical unit time period is "0"; when the When the aircraft to which the aircraft brake wheel belongs is in a landing and taxiing state in the second historical unit time period, it can be deduced that the binary value in the empty state in the second historical unit time period is "0", the binary value in the landing and taxiing state in the second historical unit time period is "1", and the binary value in the landing and stationary state in the second historical unit time period is "0"; when the aircraft to which the aircraft brake wheel belongs is in a landing and parking state in the third historical unit time period, it can be deduced that the binary value in the empty state in the third historical unit time period is "0", the binary value in the landing and taxiing state in the third historical unit time period is "0", and the binary value in the landing and stationary state in the third historical unit time period is "1".
[0074] S2. Based on the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, predict the aircraft brake wheel in the future Estimated temperature value for unit time period, where Indicates less than A positive integer.
[0075] In step S2, the variational mode decomposition algorithm (VMD) is a new signal decomposition estimation method, and its overall framework is a variational problem, which minimizes the sum of the estimated bandwidth of each mode, wherein each "mode" is assumed to be a finite bandwidth with a different center frequency. To solve this variational problem, an alternating direction multiplier method is used to continuously update each mode and its center frequency, and gradually demodulate each mode to the corresponding baseband, and finally each mode and the corresponding center frequency are extracted together; since the temperature history time series data can be regarded as a time domain signal, the variational mode decomposition algorithm can be used to decompose the temperature history time series data into multiple intrinsic modal component time series data with specific center frequencies and bandwidths. The Attention-LSTM model is an improved model that adds a layer of attention mechanism to the existing LSTM (Long Short-Term Memory, long short-term memory network) model (which is often used for time series data prediction), wherein the added attention mechanism is used to allow the LSTM model to dynamically focus on part of the information during training, so as to capture more important information and improve the performance of the model. Therefore, the temperature value can be predicted based on the variational mode decomposition algorithm and the Attention-LSTM model. Specifically, based on the temperature historical time series data, the idle state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, the aircraft brake wheel in the future can be predicted based on the variational mode decomposition algorithm and the Attention-LSTM model. The estimated temperature value of each unit time period includes but is not limited to the following steps S211 to S216.
[0076] S211. Decompose the temperature history time series data using a variational modal decomposition algorithm to obtain three modal component time series data, and make the three modal component time series data correspond one by one to the empty state, the landing gliding state, and the landing stationary state, respectively.
[0077] In step S211, since each historical temperature value in the temperature historical time series data is decomposed into three historical temperature value modal components corresponding to the three modal component time series data, each modal component time series data contains the Unit time periods and the historical temperature value modal components of each unit time period. Specifically, the parameters of the variational modal decomposition algorithm can be obtained by tuning using, but not limited to, a particle swarm algorithm. In addition, the specific method of making the three modal component time series data correspond to the empty state, the landing gliding state, and the landing stationary state can be sequential correspondence or random correspondence.
[0078] S212. For each modal component time series data in the three modal component time series data, extract corresponding multiple sample data from the corresponding time series data in the following manner: extract the sample data that are sequentially continuous in time series from the corresponding component time series data. The historical temperature value modal components of the unit time period and the corresponding state historical time series data are extracted in the unit time periods and the After unit periods and sequentially in time The values of the unit time periods are taken together as the model input items, and the corresponding component time series data are extracted in the The historical temperature value modal component of a unit period is used as the model output item, and then the model input item and the model output item are used as a sample data, wherein, Indicates greater than or equal to 2 and less than A positive integer, Indicates less than or equal to A positive integer.
[0079] In step S212, when the unit time period is 1 minute, An example is 15. For example, 5 (its value can be positively correlated with the minimum duration of the working state of the brake cooling device for cooling the aircraft brake wheel, that is, the shorter the minimum duration of the working state, The smaller the value, the smaller the value, and vice versa). In this way, the modal component of the historical temperature value from 10:01 to 10:15 (i.e., a total of 15 unit time periods) can be extracted from the modal component time series data, and the values ("1" or "0") from 10:01 to 10:15 and from 10:16 to 10:20 can be extracted from the corresponding state historical time series data, and used together as the model input items, and the modal component of the historical temperature value from 10:16 to 10:20 (i.e., a total of 5 unit time periods) can be used as the model output item, thereby obtaining a sample data.
[0080] S213. For each modal component time series data, extract multiple training sample data from the corresponding multiple sample data, and use the multiple training sample data to train the Attention-LSTM model to obtain the corresponding temperature value prediction model.
[0081] In step S213, the specific training process of the Attention-LSTM model is a prior art method; specifically, the parameters of the Attention-LSTM model can be obtained by searching using, but not limited to, a grid search algorithm or a Bayesian search algorithm. In addition, the multiple training sample data are preferably all early sample data and mid-term sample data in the multiple sample data.
[0082] S214. According to the future flight plan of the aircraft to which the aircraft brake wheel belongs, derive the time series data of the aircraft brake wheel in the air, the time series data of the landing taxiing state and the time series data of the landing stationary state, wherein the time series data of the aircraft brake wheel in the air includes but is not limited to the time series of the aircraft brake wheel in the air. future unit periods and in the The binarization of each future unit time period in the future unit time period is in the empty state value, etc., and the landing and taxiing state planning time series data includes but is not limited to the future unit periods and in the The binary landing and sliding state value of each future unit time period in the future unit time period, etc., the landing stationary state planning time series data includes but is not limited to the future unit periods and in the The binary landing static state value of each future unit time period in the future unit time period, etc. Indicates greater than or equal to A positive integer.
[0083] In step S214, since the future flight plan includes the planned state of the aircraft to which the aircraft brake wheel belongs in each future unit time period, the binary empty state value, the binary landing taxiing state value and the binary landing stationary state value in each future unit time period can be conventionally derived based on the state of the plan. For example, when the aircraft to which the aircraft brake wheel belongs is planned to be in a flight state in the first future unit time period, it can be derived that the binary empty state value in the first future unit time period is "1", the binary landing taxiing state value in the first future unit time period is "0", and the binary landing stationary state value in the first future unit time period is "0". "; when the aircraft to which the aircraft brake wheel belongs is planned to be in the landing taxiing state in the second future unit period, it can be deduced that the binary value in the empty state in the second future unit period is "0", the binary value in the landing taxiing state in the second future unit period is "1", and the binary value in the landing stationary state in the second future unit period is "0"; when the aircraft to which the aircraft brake wheel belongs is planned to be in the landing parking state in the third future unit period, it can be deduced that the binary value in the empty state in the third future unit period is "0", the binary value in the landing taxiing state in the third future unit period is "0", and the binary value in the landing stationary state in the third future unit period is "1". In addition, the future flight plan can be dynamically changed. For example, the pilot can dynamically change whether the aircraft to which the aircraft brake wheel belongs is in the flying state or the landing taxiing state in the future period through conventional flight operations, etc.
[0084] S215. For each modal component time series data, extract the most recent time series data that are consecutive in time from the corresponding component time series data. The historical temperature value modal components of the unit period and the corresponding state historical time series data are extracted in the most recent The value of the unit time period and the nearest The future that is consecutive in time after unit periods The values of the unit time periods are imported into the corresponding temperature value prediction model together, and the corresponding model output prediction new item is output.
[0085] In the step S215, continuing to refer to the example in the above step S212, if the current time is 12:00, the modal component of the historical temperature value from 11:46 to 12:00 (i.e., a total of 15 unit time periods) can be extracted from the modal component time series data, and the value from 11:46 to 12:00 (i.e., a total of 15 unit time periods) can be extracted from the corresponding state historical time series data, and the value from 12:01 to 12:05 (i.e., a total of 5 unit time periods) can be extracted from the corresponding state plan time series data, and the value from 12:01 to 12:05 (i.e., a total of 5 unit time periods) can be extracted, and imported into the temperature value prediction model together to obtain the modal component prediction result of the temperature value from 12:01 to 12:05 (i.e., a total of 5 unit time periods).
[0086] S216. Superimpose the model output prediction new item of each modal component time series data, and reconstruct the aircraft brake wheel in the future Estimated temperature value for a unit period.
[0087] In step S216, referring to the example in step S215, the estimated temperature value from 12:01 to 12:05 (i.e., a total of 5 unit time periods) can be reconstructed. In addition, since the aforementioned steps S211 to S216 are predicted by combining multiple methods, the advantages of various methods can be fully utilized to improve the accuracy and robustness of the temperature value prediction results, and generate important industry application value.
[0088] S3. According to the aircraft brake wheel in the future The estimated temperature value of the aircraft brake wheel in the future The temperature warning level for each unit period.
[0089] In step S3, since a high temperature warning is given for the change trend of the aircraft brake wheel temperature, the aircraft brake wheel is The severity of the temperature warning level for each unit period will be related to the aircraft brake wheel in the future The estimated temperature value of the unit time period is positively correlated, that is, the higher the estimated temperature value, the more serious the temperature warning level tends to be. Specifically, according to the aircraft brake wheel in the future The estimated temperature value of the aircraft brake wheel in the future The temperature warning level of each unit time period includes but is not limited to the following steps S31 to S33.
[0090] S31. According to the aircraft brake wheel in the future The estimated temperature values of each unit time period are statistically obtained to obtain the maximum estimated temperature or the average estimated temperature.
[0091] S32. According to a plurality of preset temperature value ranges corresponding to a plurality of temperature warning levels, determine a temperature value range including the estimated maximum temperature value or the estimated average temperature value from the plurality of temperature value ranges.
[0092] In the step S32, any two temperature value ranges in the multiple temperature value ranges need to ensure that there is no intersection. For example, if there are four temperature warning levels and they are respectively no warning level (its corresponding temperature value range is less than or equal to 30°C), higher temperature warning level (its corresponding temperature value range is greater than 30°C and less than or equal to 50°C), high temperature warning level (its corresponding temperature value range is greater than 50°C and less than or equal to 80°C) and ultra-high temperature warning level (its corresponding temperature value range is greater than 80°C), and the estimated maximum temperature or the estimated average temperature is 65°C, then it can be determined that a certain temperature value range is greater than 50°C and less than or equal to 80°C, and the high temperature warning level can be used as the aircraft brake wheel in the future The temperature warning level for each unit period.
[0093] S33. Using a temperature warning level corresponding to the temperature value range as the aircraft brake wheel in the future The temperature warning level for each unit period.
[0094] After step S3, in order to achieve the purpose of timely temperature control of the aircraft brake wheel, preferably, after determining that the aircraft brake wheel is in the future After the temperature warning level of a unit time period is reached, the method further includes but is not limited to the following step S4.
[0095] S4. Transmitting a preset working instruction corresponding to the temperature warning level to the brake cooling device, so as to cool down the aircraft brake wheel by controlling the working state of the brake cooling device.
[0096] In step S4, the brake cooling device may be a heat dissipation fan or a semiconductor cooling plate, etc. For the former, the working state of the brake cooling device can be controlled by changing the wind gear (for example, the larger the gear number, the lower the wind power), and for the latter, the working state of the brake cooling device can be controlled by changing the current size. For example, if the brake cooling device is a heat dissipation fan, the preset working instruction corresponding to the level without warning can be a shutdown instruction or a maintenance instruction, the preset working instruction corresponding to the higher temperature warning level can be a maintenance instruction or an adjustment instruction for indicating adjustment to the third gear wind power, the preset working instruction corresponding to the high temperature warning level can be an adjustment instruction for indicating adjustment to the second gear wind power, and the preset working instruction corresponding to the ultra-high temperature warning level can be an adjustment instruction for indicating adjustment to the first gear wind power, etc. In this way, the wind power can be increased when the warning level is serious, and the wind power can be reduced or even stopped when the warning level is not serious, and then the brake cooling device can be timely and dynamically controlled according to the warning situation to reduce energy consumption, so as to achieve the purpose of energy saving without affecting the service life of the brake device.
[0097] Therefore, based on the aircraft brake wheel temperature warning method described in the aforementioned steps S1 to S3, a new solution for aircraft brake wheel temperature prediction and high temperature warning based on temperature and state historical data is provided, that is, firstly, the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing stationary state historical time series data of the aircraft brake wheel are obtained, and then according to these data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the estimated temperature value of the aircraft brake wheel in the future period is predicted, and finally according to these estimated temperature values, the temperature warning level of the aircraft brake wheel in the future period is determined. In this way, on the one hand, by adopting a combination of multiple methods for prediction, the advantages of various methods can be fully utilized to improve the accuracy and robustness of the temperature value and high temperature warning results. On the other hand, it can also be beneficial to timely and dynamically control the brake cooling device according to the warning situation, so as to achieve the purpose of energy saving without affecting the service life of the brake device, thereby realizing the automation of the aircraft brake wheel temperature warning process and the intelligence of cooling and cooling, enhancing flight safety, and facilitating practical application and promotion.
[0098] Based on the technical solution of the first aspect, this embodiment also provides a possible design for predicting the temperature of aircraft brake wheels based on the variational mode decomposition algorithm and the Attention-LSTM model when considering the heat dissipation working state of the brake cooling device, that is, the method also includes: obtaining the heat dissipation working state historical time series data of the brake cooling device used to cool the aircraft brake wheels, wherein the heat dissipation working state historical time series data contains the working state parameter values of the brake cooling device in each unit time period. The aforementioned working state parameter values can be specific but not limited to the wind speed gear value for the heat dissipation fan or the current value for the semiconductor refrigeration plate, etc., which can be obtained by conventional records.
[0099] According to the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the aircraft brake wheel in the future is predicted. The estimated temperature value of each unit time period includes but is not limited to the following steps S221-S226.
[0100] S221. Decompose the temperature history time series data using a variational modal decomposition algorithm to obtain four modal component time series data, and make the four modal component time series data correspond one by one to the empty state, the landing and gliding state, the landing and stationary state, and the heat dissipation working state, respectively.
[0101] S222. For each modal component time series data in the four modal component time series data, extract corresponding multiple sample data from the corresponding time series data in the following manner: extract the time series data that are sequentially continuous in time from the corresponding component time series data. The historical temperature value modal components of the unit time period and the corresponding state historical time series data are extracted in the unit time periods and the After unit periods and sequentially in time The values of the unit time periods are taken together as the model input items, and the corresponding component time series data are extracted in the The historical temperature value modal component of a unit period is used as the model output item, and then the model input item and the model output item are used as a sample data, wherein, Indicates greater than or equal to 2 and less than A positive integer, Indicates less than or equal to A positive integer of ;
[0102] S223. For each modal component time series data, extract multiple training sample data from the corresponding multiple sample data, and use the multiple training sample data to perform model training on the Attention-LSTM model to obtain a corresponding temperature value prediction model;
[0103] S224. According to the future flight plan of the aircraft to which the aircraft brake wheel belongs, the planned time series data of the aircraft brake wheel in the air, the planned time series data of the landing taxiing state and the planned time series data of the landing stationary state are derived, and according to a certain working state parameter value of the most recent unit time period in the historical time series data of the heat dissipation working state, the planned time series data of the heat dissipation working state is derived, wherein the planned time series data of the air state includes but is not limited to the time series of the parameters ... future unit periods and in the The binarization of each future unit time period in the future unit time period is in the empty state value, etc., and the landing and taxiing state planning time series data includes but is not limited to the future unit periods and in the The binary landing and sliding state value of each future unit time period in the future unit time period, etc., the landing stationary state planning time series data includes but is not limited to the future unit periods and in the The binary landing static state value of each future unit time period in the future unit time period, etc., the heat dissipation working state planning time series data includes but is not limited to the future unit periods and in the Each future unit time period in a future unit time period and the working state parameter value equal to the working state parameter value, etc., Indicates greater than or equal to A positive integer.
[0104] In step S224, since the purpose of aircraft brake wheel temperature prediction and high temperature warning is to dynamically control the brake cooling device in a timely manner, it is assumed that the working state of the brake cooling device will remain unchanged in the future. That is, the heat dissipation working state planning time series data includes the future unit periods and in the The working state parameter value of each future unit time period in the future unit time periods is equal to the working state parameter value.
[0105] S225. For each modal component time series data, extract the most recent time series data that are consecutive in time from the corresponding component time series data. The historical temperature value modal components of the unit period and the corresponding state historical time series data are extracted in the most recent The value of the unit time period and the nearest The future that is consecutive in time after unit periods The values of the unit time periods are imported into the corresponding temperature value prediction model together, and the corresponding model output prediction new item is output;
[0106] S226. Superimpose the model output prediction new item of each modal component time series data, and reconstruct the aircraft brake wheel in the future Estimated temperature value for a unit period.
[0107] The specific technical details of the aforementioned steps S221 to S226 can be conventionally derived with reference to the aforementioned steps S211 to S216 and will not be repeated here.
[0108] Therefore, based on the aforementioned possible design one, by considering the heat dissipation working state of the brake cooling device, a more realistic and accurate aircraft brake wheel temperature prediction result can be obtained based on the variational mode decomposition algorithm and the Attention-LSTM model, thereby further improving the accuracy and robustness of the warning results.
[0109] Based on the technical solution of the first aspect or possible design one, this embodiment also provides a possible design two to further improve the accuracy and robustness of the warning result, that is, based on the temperature historical time series data, the empty state historical time series data, the landing taxiing state historical time series data and the landing stationary state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, it is predicted that the aircraft brake wheel will be in the future The estimated temperature value of each unit time period also includes but is not limited to the following steps S231-S237.
[0110] S231. For each modal component time series data, extract multiple test sample data from the corresponding multiple sample data, and import multiple model input items in the multiple test sample data into the corresponding temperature value prediction model, and output multiple corresponding model output prediction items, wherein the model input item of each test sample data in the multiple test sample data has the same value for all modal component time series data and is sequentially continuous in time series. The plurality of model input items correspond one-to-one to the plurality of sample data, and the plurality of model output prediction items correspond one-to-one to the plurality of model input items.
[0111] In step S231, referring to the example of step S212, if for the modal component time series data A, the first test sample data a having a model input item from 10:01 to 10:15 (i.e., a total of 15 unit time periods), the second test sample data a having a model input item from 10:06 to 10:20 (i.e., a total of 15 unit time periods), and the model input item from 10:11 to 10:25 (i.e., a total of 15 unit time periods) have been extracted from the corresponding multiple sample data. 15 unit time periods) etc., then for the modal component time series data B, it is also necessary to extract the first test sample data b with a model input item from 10:01 to 10:15 (i.e., a total of 15 unit time periods), the second test sample data b with a model input item from 10:06 to 10:20 (i.e., a total of 15 unit time periods), and the third test sample data a with a model input item from 10:11 to 10:25 (i.e., a total of 15 unit time periods) from the corresponding multiple sample data. For the modal component time series data C, it is also necessary to extract from the corresponding multiple sample data the first test sample data c with a model input item from 10:01 to 10:15 (i.e., a total of 15 unit time periods), the second test sample data c with a model input item from 10:06 to 10:20 (i.e., a total of 15 unit time periods), and the third test sample data c with a model input item from 10:11 to 10:25 (i.e., a total of 15 unit time periods). Sample data c, etc.; for the modal component time series data D, it is also necessary to extract from the corresponding multiple sample data the first test sample data d whose model input items are from 10:01 to 10:15 (i.e., a total of 15 unit time periods), the second test sample data d whose model input items are from 10:06 to 10:20 (i.e., a total of 15 unit time periods), and the third test sample data d whose model input items are from 10:11 to 10:25 (i.e., a total of 15 unit time periods). Since the model output items contain The historical temperature value modal component of each unit period, so the model output prediction item will also include In addition, the plurality of test sample data are preferably all late sample data in the plurality of sample data.
[0112] S232. The same as described in each group unit time periods, superimpose the time series data of each modal component and the corresponding model output prediction item, and reconstruct the time series data that is located after the corresponding time period and is sequentially continuous in time series. The estimated temperature value of each unit period, and the values that are located after the corresponding period and are consecutive in time sequence The historical temperature values of each unit period are subtracted one by one from the historical temperature values of the corresponding period and consecutive in time sequence. The estimated temperature value of each unit period is obtained after the corresponding period and is sequentially continuous in time sequence. The temperature error value per unit time period.
[0113] In the step S232, continuing to refer to the examples in the aforementioned steps S212 and S231, for the period from 10:01 to 10:15 (i.e., a total of 15 unit time periods), the modal component time series data A and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:16 to 10:20 predicted based on the model input item in the first test sample data a), the modal component time series data B and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:16 to 10:20 predicted based on the model input item in the first test sample data b), and the modal component time series data C and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:16 to 10:20 predicted based on the model input item in the first test sample data b) can be superimposed. The model input item in the sample data c predicts the modal component prediction result of the temperature value from 10:16 to 10:20) and the modal component time series data D and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:16 to 10:20 predicted based on the model input item in the first test sample data d), reconstructs the estimated temperature value from 10:16 to 10:20 (i.e., a total of 5 unit time periods), and makes the historical temperature values from 10:16 to 10:20 (i.e., a total of 5 historical temperature values) subtract the estimated temperature value from 10:16 to 10:20 (i.e., a total of 5 estimated temperature values) one by one, and the temperature from 10:16 to 10:20 can be calculated. temperature error value (i.e., a total of 5 temperature error values); for the period from 10:06 to 10:20 (i.e., a total of 15 unit time periods), the modal component time series data A and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:21 to 10:25 predicted based on the model input item in the second test sample data a), the modal component time series data B and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:21 to 10:25 predicted based on the model input item in the second test sample data b), the modal component time series data C and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:21 to 10:25 predicted based on the model input item in the second test sample data c), The predicted results of the modal components of the temperature values from 10:21 to 10:25) and the modal component time series data D and the corresponding model output predicted items (i.e., the predicted results of the modal components of the temperature values from 10:21 to 10:25 and obtained based on the model input items in the second test sample data d), the estimated temperature values from 10:21 to 10:25 (i.e., a total of 5 unit time periods) are reconstructed, and the estimated temperature values from 10:21 to 10:25 (i.e., a total of 5 fixed temperature values) are subtracted one by one from the historical temperature values from 10:21 to 10:25 (i.e., a total of 5 historical temperature values), and the temperature error values from 10:21 to 10:25 (i.e., a total of 5 temperature error values) can be calculated;For the period from 10:11 to 10:25 (i.e., a total of 15 unit time periods), the modal component time series data A and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:26 to 10:30 predicted based on the model input item in the third test sample data a), the modal component time series data B and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:26 to 10:30 predicted based on the model input item in the third test sample data b), the modal component time series data C and the corresponding model output prediction item (i.e., the modal component prediction result of the temperature value from 10:26 to 10:30 predicted based on the model input item in the third test sample data c) are superimposed. 10:26 to 10:30 (i.e., a total of 5 unit periods) are reconstructed based on the model input item prediction result of the temperature value modal component from 10:26 to 10:30) and the modal component time series data D and the corresponding model output prediction item (i.e., the model input item prediction result of the temperature value modal component from 10:26 to 10:30 is obtained based on the third test sample data d), and the estimated temperature value from 10:26 to 10:30 (i.e., a total of 5 unit periods) is subtracted from the historical temperature values from 10:26 to 10:30 (i.e., a total of 5 historical temperature values) one by one, and the temperature error value from 10:26 to 10:30 (i.e., a total of 5 temperature error values) can be calculated. ;
[0114] S233. According to the same After unit periods and sequentially in time The temperature error value of each unit time period is constructed to obtain the temperature error time series data, wherein the temperature error time series data contains the temperature error values that are consecutive in time series. unit time period and in the The temperature error value of each unit time period in the unit time period is Represents a positive integer greater than or equal to 3.
[0115] In the step S233, the same After unit periods and sequentially in time The temperature error values of the unit time periods are obtained to obtain the temperature error time series data. In addition, during the sorting process, if there are multiple temperature error values in a unit time period, the average value of the multiple temperature error values can be used as the final temperature error value of the unit time period.
[0116] S234. Extract a plurality of new sample data from the temperature error time series data in the following manner: extract the temperature error time series data that are sequentially continuous in time series. The temperature error value of each unit period is used as a new model input item, and the temperature error time series data is extracted from the After unit periods and sequentially in time The temperature error value of each unit period is used as a new model output item, and then the model input new item and the model output new item are used as a new sample data, wherein, , .
[0117] In the step S234, continuing to refer to the examples in the aforementioned step S212 and steps S231 to S232, the temperature error values from 10:16 to 10:30 (i.e., a total of 15 unit time periods) can be extracted from the temperature error time series data as new model input items, and the temperature error values from 10:31 to 10:35 (i.e., a total of 5 unit time periods) can be extracted as new model output items, thereby obtaining a new sample data.
[0118] S235. Use the multiple new sample data to perform model training on the LSTM model to obtain a temperature error prediction model.
[0119] In step S235, the specific training process of the LSTM model is a prior art method; specifically, the parameters of the LSTM model can be obtained by searching using, but not limited to, a grid search algorithm or a Bayesian search algorithm.
[0120] S236. Extract the most recent The temperature error value of each unit period is imported into the temperature error prediction model to obtain the future The predicted value of temperature error for each unit period.
[0121] In the step S236, since the plurality of test sample data are preferably all late sample data in the plurality of sample data, it is ensured that the most recent sample data can be extracted from the temperature error time series data. Continuing with the example in the aforementioned step S215, the temperature error values from 11:46 to 12:00 (i.e., a total of 15 unit time periods) can be extracted from the temperature error time series data and imported into the temperature error prediction model to obtain the temperature error prediction value from 12:01 to 12:05 (i.e., a total of 5 unit time periods).
[0122] S237. Make the future The estimated temperature values of each unit period are added to the future The temperature error prediction value of the aircraft brake wheel in the future The final estimated temperature value for the unit period.
[0123] In step S237, continuing to refer to the examples in the above-mentioned steps S212 and S236, the estimated temperature values from 12:01 to 12:05 (i.e., a total of 5 unit time periods) can be added one-to-one to the temperature error prediction values from 12:01 to 12:05 (i.e., a total of 5 unit time periods), thereby calculating the final estimated temperature value from 12:01 to 12:05 (i.e., a total of 5 unit time periods) .
[0124] Therefore, based on the aforementioned possible design 2, the accuracy and robustness of the warning results can be further improved by constructing temperature error time series data, training a temperature error prediction model based on this data, and applying the temperature error prediction model to correct the estimated temperature values of the aircraft brake wheels in the future period.
[0125] like Figure 2 As shown, the second aspect of this embodiment provides a computer device for executing the aircraft brake wheel temperature warning method as described in the first aspect or any possible design in the first aspect, including a memory, a processor and a transceiver that are sequentially connected in communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the aircraft brake wheel temperature warning method as described in the first aspect or any possible design in the first aspect. For example, the memory may include, but is not limited to, a random access memory (Random-Access Memory, RAM), a read-only memory (Read-Only Memory, ROM), a flash memory (FlashMemory), a first-in-first-out memory (First Input First Output, FIFO) and / or a first-in-last-out memory (First Input Last Output, FILO), etc.; the processor may include, but is not limited to, a microprocessor of the STM32F105 series. In addition, the computer device may also include, but is not limited to, a power module, a display screen and other necessary components.
[0126] The working process, working details and technical effects of the aforementioned computer device provided in the second aspect of this embodiment can be referred to the aircraft brake wheel temperature warning method described in the first aspect or any possible design of the first aspect, and will not be described in detail here.
[0127] The third aspect of this embodiment provides a computer-readable storage medium storing instructions for the aircraft brake wheel temperature warning method as described in the first aspect or any possible design of the first aspect, that is, the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the aircraft brake wheel temperature warning method as described in the first aspect or any possible design of the first aspect is executed. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to computer-readable storage media such as floppy disks, optical disks, hard disks, flash memory, USB flash drives and / or memory sticks, and the computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices.
[0128] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the third aspect of this embodiment can be referred to the aircraft brake wheel temperature warning method as described in the first aspect or any possible design in the first aspect, and will not be described in detail here.
[0129] A fourth aspect of the present embodiment provides a computer program product, including a computer program or an instruction, wherein when the computer program or the instruction is executed by a computer, the aircraft brake wheel temperature warning method as described in the first aspect or any possible design of the first aspect is implemented. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
[0130] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A method for early warning of aircraft brake wheel temperature based on data processing, It is characterized in that include: Obtain the temperature history time series data of the aircraft brake wheel, the history time series data of the empty state, the history time series data of the landing taxiing state and the history time series data of the landing static state, wherein the temperature history time series data contains the unit time period and in the the historical temperature value of each unit time period in the unit time period, the empty state historical time series data contains the binary empty state value in each unit time period, the binary empty state value is "1" indicating that the aircraft brake wheel is in the empty state in the corresponding unit time period, and the binary empty state value is "0" indicating that the aircraft brake wheel is not in the empty state in the corresponding unit time period, the landing taxiing state historical time series data contains the binary landing taxiing state value in each unit time period, the binary landing taxiing state value is "1" indicating that the aircraft brake wheel is in the landing taxiing state in the corresponding unit time period, and the binary landing taxiing state value is "0" indicating that the aircraft brake wheel is in the non-landing taxiing state in the corresponding unit time period, the landing static state historical time series data contains the binary landing static state value in each unit time period, the binary landing static state value is "1" indicating that the aircraft brake wheel is in the landing static state in the corresponding unit time period, and the binary landing static state value is "0" indicating that the aircraft brake wheel is in the non-landing static state in the corresponding unit time period, represents a positive integer greater than or equal to 3; According to the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the aircraft brake wheel in the future is predicted. The estimated temperature value of a unit time period specifically includes: using a variational mode decomposition algorithm to decompose the temperature history time series data to obtain three modal component time series data, and making the three modal component time series data correspond to the empty state, the landing and sliding state, and the landing and stationary state respectively; for each modal component time series data in the three modal component time series data, extracting corresponding multiple sample data from the corresponding time series data in the following manner: extracting the corresponding component time series data that are consecutive in time series The historical temperature value modal components of the unit time period and the corresponding state historical time series data are extracted in the unit time periods and the After unit periods and sequentially in time The values of the unit time periods are taken together as the model input items, and the corresponding component time series data are extracted in the The historical temperature value modal component of a unit period is used as the model output item, and then the model input item and the model output item are used as a sample data, wherein, Indicates greater than or equal to 2 and less than A positive integer, Indicates less than or equal to A positive integer; for each modal component time series data, extract multiple training sample data from the corresponding multiple sample data, and use the multiple training sample data to train the Attention-LSTM model to obtain a corresponding temperature value prediction model; according to the future flight plan of the aircraft to which the aircraft brake wheel belongs, derive the empty state planned time series data, the landing taxiing state planned time series data and the landing stationary state planned time series data of the aircraft brake wheel, wherein the empty state planned time series data contains sequentially in time series future unit periods and in the The binarization of each future unit time period in the future unit time period is in the empty state value, and the landing and taxiing state planning time series data contains the future unit periods and in the The binary landing and sliding state value of each future unit time period in the future unit time periods, the landing stationary state planning time series data contains the future unit periods and in the The binary ground stationary state value of each future unit time period in the future unit time periods, Indicates greater than or equal to A positive integer; for each modal component time series data, extract the most recent consecutive time series data from the corresponding component time series data The historical temperature value modal components of the unit period and the corresponding state historical time series data are extracted in the most recent The value of the unit time period and the nearest The future that is consecutive in time after unit periods The values of the unit time periods are imported into the corresponding temperature value prediction model together, and the corresponding model output prediction new items are output; the model output prediction new items of the time series data of each modal component are superimposed, and the aircraft brake wheel is reconstructed in the future Estimated temperature value for a unit period; According to the aircraft brake wheel in the future The estimated temperature value of the aircraft brake wheel in the future The temperature warning level for each unit period.
2. The aircraft brake wheel temperature early warning method according to claim 1, It is characterized in that According to the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the aircraft brake wheel in the future is predicted. Estimated temperature values for each unit period, including: For each of the modal component time series data, a plurality of test sample data are extracted from the corresponding plurality of sample data, and a plurality of model input items in the plurality of test sample data are introduced into the corresponding temperature value prediction model, and a plurality of corresponding model output prediction items are output, wherein the model input item of each of the plurality of test sample data has the same value for all the modal component time series data and is sequentially continuous in time series. unit time period, the plurality of model input items correspond one-to-one to the plurality of sample data, and the plurality of model output prediction items correspond one-to-one to the plurality of model input items; The same unit time periods, superimpose the time series data of each modal component and the corresponding model output prediction item, and reconstruct the time series data that is located after the corresponding time period and is sequentially continuous in time series. The estimated temperature value of each unit period, and the values that are located after the corresponding period and are consecutive in time sequence The historical temperature values of each unit period are subtracted one by one from the historical temperature values of the corresponding period and consecutive in time sequence. The estimated temperature value of each unit period is obtained after the corresponding period and is sequentially continuous in time sequence. Temperature error value per unit time period; According to the same After unit periods and sequentially in time The temperature error value of each unit time period is constructed to obtain the temperature error time series data, wherein the temperature error time series data contains the temperature error values that are consecutive in time series. unit time period and in the The temperature error value of each unit time period in the unit time period is represents a positive integer greater than or equal to 3; A plurality of new sample data are extracted from the temperature error time series data in the following manner: extracting the temperature error time series data that are sequentially continuous in time series. The temperature error value of each unit period is used as a new model input item, and the temperature error time series data is extracted from the After unit periods and sequentially in time The temperature error value of each unit period is used as a new model output item, and then the model input new item and the model output new item are used as a new sample data, wherein, , ; Using the multiple new sample data to perform model training on the LSTM model to obtain a temperature error prediction model; Extract the most recent The temperature error value of each unit period is imported into the temperature error prediction model to obtain the future Temperature error prediction value for each unit period; Make the future The estimated temperature values of each unit period are added to the future The temperature error prediction value of the aircraft brake wheel in the future The final estimated temperature value for the unit period.
3. The aircraft brake wheel temperature early warning method according to claim 1, It is characterized in that According to the aircraft brake wheel in the future The estimated temperature value of the aircraft brake wheel in the future The temperature warning level for each unit period includes: According to the aircraft brake wheel in the future The estimated temperature values of the unit time periods are statistically obtained to obtain the maximum estimated temperature or the average estimated temperature; According to a plurality of temperature value ranges that are preset and correspond one to one with a plurality of temperature warning levels, determining a temperature value range that includes the estimated maximum temperature value or the estimated average temperature value from the plurality of temperature value ranges; A temperature warning level corresponding to the temperature value range is used as the aircraft brake wheel in the future The temperature warning level for each unit period.
4. The aircraft brake wheel temperature early warning method according to claim 1, It is characterized in that In determining the future After the temperature warning level of a unit period is determined, the method further comprises: A preset working instruction corresponding to the temperature warning level is transmitted to the brake cooling device, so as to cool down the aircraft brake wheel by controlling the working state of the brake cooling device.
5. The aircraft brake wheel temperature early warning method according to claim 4, It is characterized in that The brake cooling device includes a heat dissipation fan and / or a semiconductor cooling plate.
6. The aircraft brake wheel temperature early warning method according to claim 1, It is characterized in that Obtain the temperature history time series data of aircraft brake wheels, including: Acquiring raw temperature data periodically collected by a temperature sensor installed in the aircraft brake wheel; The original temperature data is preprocessed to obtain the temperature history time series data of the aircraft brake wheel, wherein the preprocessing includes data cleaning processing, data interpolation processing, data abnormality processing and / or data normalization processing, and the temperature history time series data contains the temperature history data which are sequentially continuous in time series. unit time period and in the The historical temperature values of each unit period in the unit period, Represents a positive integer greater than or equal to 3.
7. The aircraft brake wheel temperature early warning method according to claim 1, It is characterized in that Obtain the historical time series data of the aircraft brake wheels in the air, the landing taxiing state and the landing stationary state, including: According to the flight history record of the aircraft to which the aircraft brake wheel belongs, the historical time series data of the aircraft brake wheel in the air, the historical time series data of the landing taxiing state and the historical time series data of the landing static state are derived, wherein the historical time series data of the aircraft brake wheel in the air contains the following data: unit time period and in the The binary empty state value of each unit time period in the unit time period is "1", indicating that the aircraft brake wheel is in the empty state in the corresponding unit time period, and the binary empty state value is "0", indicating that the aircraft brake wheel is not in the empty state in the corresponding unit time period. The landing taxiing state historical time series data contains the binary landing taxiing state value in each unit time period, the binary landing taxiing state value is "1", indicating that the aircraft brake wheel is in the landing taxiing state in the corresponding unit time period, and the binary landing taxiing state value is "0", indicating that the aircraft brake wheel is in the non-landing taxiing state in the corresponding unit time period. The landing static state historical time series data contains the binary landing static state value in each unit time period, the binary landing static state value is "1", indicating that the aircraft brake wheel is in the landing static state in the corresponding unit time period, and the binary landing static state value is "0", indicating that the aircraft brake wheel is in the non-landing static state in the corresponding unit time period. Represents a positive integer greater than or equal to 3.
8. The aircraft brake wheel temperature early warning method according to claim 1, It is characterized in that The parameters of the variational mode decomposition algorithm are tuned using a particle swarm algorithm, and the parameters of the Attention-LSTM model are searched using a grid search algorithm or a Bayesian search algorithm.
9. A method for early warning of aircraft brake wheel temperature based on data processing, It is characterized in that include: Obtain the temperature history time series data of the aircraft brake wheel, the history time series data of the empty state, the history time series data of the landing taxiing state and the history time series data of the landing static state, wherein the temperature history time series data contains the unit time period and in the the historical temperature value of each unit time period in the unit time period, the empty state historical time series data contains the binary empty state value in each unit time period, the binary empty state value is "1" indicating that the aircraft brake wheel is in the empty state in the corresponding unit time period, and the binary empty state value is "0" indicating that the aircraft brake wheel is not in the empty state in the corresponding unit time period, the landing taxiing state historical time series data contains the binary landing taxiing state value in each unit time period, the binary landing taxiing state value is "1" indicating that the aircraft brake wheel is in the landing taxiing state in the corresponding unit time period, and the binary landing taxiing state value is "0" indicating that the aircraft brake wheel is in the non-landing taxiing state in the corresponding unit time period, the landing static state historical time series data contains the binary landing static state value in each unit time period, the binary landing static state value is "1" indicating that the aircraft brake wheel is in the landing static state in the corresponding unit time period, and the binary landing static state value is "0" indicating that the aircraft brake wheel is in the non-landing static state in the corresponding unit time period, represents a positive integer greater than or equal to 3; Acquire historical time series data of the heat dissipation working state of a brake cooling device used to cool the aircraft brake wheel, wherein the historical time series data of the heat dissipation working state contains working state parameter values of the brake cooling device in each unit time period; According to the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the aircraft brake wheel in the future is predicted. The estimated temperature value of a unit time period specifically includes: using a variational mode decomposition algorithm to decompose the temperature history time series data to obtain four modal component time series data, and making the four modal component time series data correspond to the empty state, the landing and sliding state, the landing and static state and the heat dissipation working state respectively; for each modal component time series data in the four modal component time series data, extracting corresponding multiple sample data from the corresponding time series data in the following manner: extracting the modal component time series data that are consecutive in time series from the corresponding component time series data The historical temperature value modal components of the unit time period and the corresponding state historical time series data are extracted in the unit time periods and the After unit periods and sequentially in time The values of the unit time periods are taken together as the model input items, and the corresponding component time series data are extracted in the The historical temperature value modal component of a unit period is used as the model output item, and then the model input item and the model output item are used as a sample data, wherein, Indicates greater than or equal to 2 and less than A positive integer, Indicates less than or equal to A positive integer; for each modal component time series data, extract multiple training sample data from the corresponding multiple sample data, and use the multiple training sample data to train the Attention-LSTM model to obtain a corresponding temperature value prediction model; according to the future flight plan of the aircraft to which the aircraft brake wheel belongs, derive the empty state planned time series data, the landing taxiing state planned time series data and the landing stationary state planned time series data of the aircraft brake wheel, and according to a working state parameter value of the most recent unit time period in the heat dissipation working state historical time series data, derive the heat dissipation working state planned time series data, wherein the empty state planned time series data contains sequentially in time series. future unit periods and in the The binarization of each future unit time period in the future unit time period is in the empty state value, and the landing and taxiing state planning time series data contains the future unit periods and in the The binary landing and sliding state value of each future unit time period in the future unit time periods, the landing stationary state planning time series data contains the future unit periods and in the The binary landing static state value of each future unit time period in the future unit time period, the heat dissipation working state planning time series data contains the future unit periods and in the each future unit time period in a future unit time period and equal to the working state parameter value of the certain working state parameter value, Indicates greater than or equal to A positive integer; for each modal component time series data, extract the most recent consecutive time series data from the corresponding component time series data The historical temperature value modal components of the unit period and the corresponding state historical time series data are extracted in the most recent The value of the unit time period and the nearest The future that is consecutive in time after unit periods The values of the unit time periods are imported into the corresponding temperature value prediction model together, and the corresponding model output prediction new items are output; the model output prediction new items of the time series data of each modal component are superimposed, and the aircraft brake wheel is reconstructed in the future Estimated temperature value for a unit period; According to the aircraft brake wheel in the future The estimated temperature value of the aircraft brake wheel in the future The temperature warning level for each unit period.
10. The aircraft brake wheel temperature early warning method according to claim 9, It is characterized in that According to the temperature historical time series data, the air state historical time series data, the landing taxiing state historical time series data and the landing static state historical time series data, based on the variational mode decomposition algorithm and the Attention-LSTM model, the aircraft brake wheel in the future is predicted. Estimated temperature values for each unit period, including: For each of the modal component time series data, a plurality of test sample data are extracted from the corresponding plurality of sample data, and a plurality of model input items in the plurality of test sample data are introduced into the corresponding temperature value prediction model, and a plurality of corresponding model output prediction items are output, wherein the model input item of each of the plurality of test sample data has the same value for all the modal component time series data and is sequentially continuous in time series. unit time period, the plurality of model input items correspond one-to-one to the plurality of sample data, and the plurality of model output prediction items correspond one-to-one to the plurality of model input items; The same unit time periods, superimpose the time series data of each modal component and the corresponding model output prediction item, and reconstruct the time series data that is located after the corresponding time period and is sequentially continuous in time series. The estimated temperature value of each unit period, and the values that are located after the corresponding period and are consecutive in time sequence The historical temperature values of each unit period are subtracted one by one from the historical temperature values of the corresponding period and consecutive in time sequence. The estimated temperature value of each unit period is obtained after the corresponding period and is sequentially continuous in time sequence. Temperature error value per unit time period; According to the same After unit periods and sequentially in time The temperature error value of each unit time period is constructed to obtain the temperature error time series data, wherein the temperature error time series data contains the temperature error values that are consecutive in time series. unit time period and in the The temperature error value of each unit time period in the unit time period is represents a positive integer greater than or equal to 3; A plurality of new sample data are extracted from the temperature error time series data in the following manner: extracting the temperature error time series data that are sequentially continuous in time series. The temperature error value of each unit period is used as a new model input item, and the temperature error time series data is extracted from the After unit periods and sequentially in time The temperature error value of each unit period is used as a new model output item, and then the model input new item and the model output new item are used as a new sample data, wherein, , ; Using the multiple new sample data to perform model training on the LSTM model to obtain a temperature error prediction model; Extract the most recent The temperature error value of each unit period is imported into the temperature error prediction model to obtain the future Temperature error prediction value for each unit period; Make the future The estimated temperature values of each unit period are added to the future The temperature error prediction value of the aircraft brake wheel in the future The final estimated temperature value for the unit period.
