CN117763976A - method and device for predicting lubricating oil quantity of aero-engine and computer equipment - Google Patents
method and device for predicting lubricating oil quantity of aero-engine and computer equipment Download PDFInfo
- Publication number
- CN117763976A CN117763976A CN202410194385.3A CN202410194385A CN117763976A CN 117763976 A CN117763976 A CN 117763976A CN 202410194385 A CN202410194385 A CN 202410194385A CN 117763976 A CN117763976 A CN 117763976A
- Authority
- CN
- China
- Prior art keywords
- lubricating oil
- sub
- models
- oil quantity
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000010687 lubricating oil Substances 0.000 title claims abstract description 131
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000007621 cluster analysis Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 80
- 239000011159 matrix material Substances 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 239000003921 oil Substances 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 10
- 238000005461 lubrication Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 4
- 230000001050 lubricating effect Effects 0.000 abstract description 7
- 238000012216 screening Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004138 cluster model Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the field of aero-engines, in particular to a method, a device and computer equipment for predicting the lubricating oil quantity of an aero-engine, which are used for screening a plurality of sub-models in a model cluster with better performance through a cluster analysis method, extracting a plurality of target sub-models, constructing a target lubricating oil quantity prediction model of the aero-engine, optimizing the performance of an integral model, and improving the accuracy and efficiency of the lubricating oil quantity prediction of the aero-engine, so that an aero-engine lubricating system is monitored more accurately and effectively.
Description
Technical Field
The invention relates to the field of aero-engines, in particular to a method and a device for predicting the lubricating oil quantity of an aero-engine, computer equipment and a storage medium.
Background
the lubricating system of the aero-engine plays a key role, such as that the engine is stopped due to faults, so that the flight safety is seriously influenced, and the fault monitoring of the lubricating system is still mostly based on experience and qualitative analysis at present, so that the accuracy and quantification are not enough.
the traditional monitoring means comprise an over-low oil quantity alarm of the aeroengine, ground manual monitoring and double-sided engine aeroengine oil quantity difference comparison, but the problems that the real-time performance, the analysis is time-consuming, the double-sided engine is difficult to solve and the like cannot be achieved. Therefore, the use of the monitored value of the characteristic quantity of the lubricating oil system to simply and effectively predict the health condition of the lubricating oil system of the engine is a problem to be studied urgently.
Disclosure of Invention
Based on the above, the invention aims to provide an aeroengine lubricating oil quantity prediction method, device, computer equipment and storage medium, which are used for screening a plurality of sub-models in an aeroengine lubricating oil quantity prediction model through a cluster analysis method, reserving the sub-models in a better-performing model cluster, extracting a plurality of target sub-models, constructing a target aeroengine lubricating oil quantity prediction model, optimizing the performance of an integral model, and improving the accuracy and efficiency of aeroengine lubricating oil quantity prediction, so that an aeroengine lubricating system is monitored more accurately and effectively.
in a first aspect, an embodiment of the present application provides a method for predicting an amount of lubricating oil for an aeroengine, including the following steps:
obtaining a flight parameter time sequence data training set and an aeroengine lubricating oil mass prediction model to be trained, wherein the flight parameter time sequence data training set comprises a plurality of flight parameter time sequence data, and the aeroengine lubricating oil mass prediction model comprises a plurality of sub-models;
Constructing a plurality of sample training sets corresponding to the sub-models according to the time sequence data training set of the flight parameters, inputting the plurality of sample training sets into the corresponding sub-models for training, and obtaining loss values of the plurality of sub-models;
Constructing a plurality of model clusters by adopting a clustering analysis method according to the loss values of the plurality of sub-models, wherein the model clusters comprise a plurality of sub-models of the same clustering label; according to the loss values of a plurality of sub-models in a plurality of model clusters, extracting a plurality of target sub-models from a plurality of model clusters, and constructing a target aero-engine lubricating oil quantity prediction model;
And responding to a prediction instruction, obtaining time sequence data of flight parameters to be predicted, inputting the time sequence data to be predicted into the target aircraft engine lubricating oil quantity prediction model, obtaining a plurality of aircraft engine lubricating oil quantity predicted values output by the target sub-model, and carrying out average processing on the aircraft engine lubricating oil quantity predicted values output by the target sub-model to obtain an aircraft engine lubricating oil quantity predicted average value as an aircraft engine lubricating oil quantity predicted result of the time sequence data of the flight parameters to be predicted.
in a second aspect, an embodiment of the present application provides an apparatus for predicting an amount of lubricating oil for an aeroengine, including:
The system comprises a data acquisition module, a data prediction module and a data processing module, wherein the data acquisition module is used for acquiring a flight parameter time sequence data training set and an aeroengine lubricating oil quantity prediction model to be trained, the flight parameter time sequence data training set comprises a plurality of flight parameter time sequence data, and the aeroengine lubricating oil quantity prediction model comprises a plurality of sub-models;
The sub-model training module is used for constructing a plurality of sample training sets corresponding to the sub-models according to the time sequence data training sets of the flight parameters, inputting the plurality of sample training sets into the corresponding sub-models for training, and obtaining loss values of the plurality of sub-models;
The sub-model extraction module is used for constructing a plurality of model clusters by adopting a cluster analysis method according to the loss values of the plurality of sub-models, wherein the model clusters comprise a plurality of sub-models with the same cluster labels; according to the loss values of a plurality of sub-models in a plurality of model clusters, extracting a plurality of target sub-models from a plurality of model clusters, and constructing a target aero-engine lubricating oil quantity prediction model;
The prediction module is used for responding to a prediction instruction, obtaining time sequence data of flight parameters to be predicted, inputting the time sequence data to be predicted into the target aircraft engine lubricating oil quantity prediction model, obtaining a plurality of aircraft engine lubricating oil quantity prediction values output by the target submodel, and carrying out average processing on the aircraft engine lubricating oil quantity prediction values output by the target submodel to obtain an aircraft engine lubricating oil quantity prediction average value which is used as an aircraft engine lubricating oil quantity prediction result of the time sequence data of the flight parameters to be predicted.
In a third aspect, an embodiment of the present application provides a computer apparatus, including: a processor, a memory, and a computer program stored on the memory and executable on the processor; the computer program when executed by the processor implements the steps of the method for predicting the amount of lubricating oil for an aircraft engine according to the first aspect.
In a fourth aspect, an embodiment of the present application provides a storage medium storing a computer program, which when executed by a processor implements the steps of the method for predicting the amount of lubricating oil for an aero-engine according to the first aspect.
In the embodiment of the application, a method, a device, computer equipment and a storage medium for predicting the lubricating oil quantity of an aero-engine are provided, a plurality of sub-models in a model cluster with better performance are reserved by screening the sub-models in the lubricating oil quantity prediction model of the aero-engine through a cluster analysis method, a plurality of target sub-models are extracted, a target lubricating oil quantity prediction model of the aero-engine is constructed, the performance of an integral model is optimized, the accuracy and the efficiency of the lubricating oil quantity prediction of the aero-engine are improved, and therefore an aero-engine lubricating system is monitored more accurately and effectively.
For a better understanding and implementation, the present invention is described in detail below with reference to the drawings.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting the amount of lubricating oil for an aircraft engine according to an embodiment of the present application;
fig. 2 is a schematic flow chart of S2 in the method for predicting the amount of lubricating oil of an aero-engine according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of S21 in an aero-engine lubrication oil quantity prediction method according to an embodiment of the present application;
fig. 4 is a schematic flow chart of S3 in the method for predicting the amount of lubricating oil of an aero-engine according to an embodiment of the present application;
Fig. 5 is a schematic structural diagram of an aero-engine lubricating oil quantity prediction device according to an embodiment of the present application;
Fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the application. The word "if"/"if" as used herein may be interpreted as "at … …" or "at … …" or "in response to a determination", depending on the context.
The data transmitting end can be a computer device or a mobile terminal device, and is used for establishing network connection with the data receiving end, encoding data information transmitted to the data receiving end and analyzing the data information transmitted from the data receiving end.
the data receiving end can be a computer device or a mobile terminal device, and is used for establishing network connection with the data sending end, encoding data information sent to the data sending end, and analyzing the data information sent from the data sending end.
referring to fig. 1, fig. 1 is a flow chart of a method for predicting an amount of lubricating oil of an aero-engine according to an embodiment of the present application, the method includes the following steps:
S1: and obtaining a flight parameter time sequence data training set and an aeroengine lubricating oil quantity prediction model to be trained.
The main body of execution of the method for predicting the amount of lubricating oil for an aircraft engine according to the present application is a prediction device (hereinafter referred to as a prediction device) for the method for predicting the amount of lubricating oil for an aircraft engine. In an alternative embodiment, the prediction device may be a computer device, may be a server, or may be a server cluster formed by combining multiple computer devices.
In this embodiment, the prediction device obtains a flight parameter time series data training set and an aeroengine lubricating oil mass prediction model to be trained, where the flight parameter time series data training set includes engine high-pressure rotor rotational speed time series data, flight attitude time series data, flight altitude time series data, and ambient temperature time series data, and the aeroengine lubricating oil mass prediction model includes a plurality of sub-models.
specifically, the prediction device collects recording data related to the oil consumption rate of the aircraft engine in the flight process of the aircraft, wherein the recording data is that QAR data recording device is adopted to read various data from an aircraft bus (such as ARINC 429 bus or AFDX bus), data collected every second are stored in a frame-subframe-subslot mode according to a preset recording configuration table, binary data streams are converted into engineering values recorded by a sensor by adopting decoding software and serve as the recording data, the data related to the oil consumption rate of the aircraft is extracted according to the recording data, the data are converted into time sequence data, and the flight parameter time sequence data training set is constructed.
S2: according to the time sequence data training set of the flight parameters, a plurality of sample training sets corresponding to the sub-models are constructed, the plurality of sample training sets are input into the corresponding sub-models for training, and loss values of the plurality of sub-models are obtained.
In this embodiment, the prediction device constructs a plurality of sample training sets corresponding to the sub-models by adopting a Bootstrap (self-sampling) method according to the time-series data training set of the flight parameters, and inputs the plurality of sample training sets into the corresponding sub-models to train, thereby obtaining loss values of the plurality of sub-models.
referring to fig. 2, fig. 2 is a schematic flow chart of step S2 in the method for predicting the lubrication oil amount of an aero-engine according to an embodiment of the application, including steps S21 to S22, specifically including the following steps:
S21: and inputting the plurality of sample training sets into the corresponding sub-models to obtain a plurality of aeroengine lubricating oil quantity predicted values corresponding to the plurality of sample training sets output by the plurality of sub-models.
The sub-model is a random forest model formed by decision trees, and specifically, transformer Encoder is adopted as the sub-model by prediction equipment.
In this embodiment, the prediction device inputs a plurality of sample training sets to the corresponding sub-models, and obtains the predicted values of the lubricating oil amounts of the aero-engine corresponding to the plurality of sample training sets output by the plurality of sub-models.
referring to fig. 3, fig. 3 is a schematic flow chart of step S21 in the method for predicting the lubrication oil amount of an aero-engine according to an embodiment of the present application, including steps S211 to S213, which specifically include the following steps:
S211: and carrying out space coding on the sample training set to obtain a plurality of space matrixes of the sample training set.
In order to avoid feature collision, in this embodiment, the prediction device spatially encodes the sample training set to obtain a plurality of spatial matrices of the sample training set, where the spatial matrices include a first spatial matrix, a second spatial matrix, and a third spatial matrix, and specifically includes:
In the method, in the process of the invention,QIn the case of a first spatial matrix,Kin the case of a second spatial matrix,VIn the case of a third spatial matrix,XIn order to provide a training set of samples,、/>、/>The third weight matrix, the fourth weight matrix and the fifth weight matrix are respectively adopted.
s212: and respectively taking the space matrix of the sample training set as input data of a plurality of self-attention heads of the self-attention module, and obtaining attention characteristic values according to a preset self-attention extraction algorithm.
the attention extraction algorithm is as follows:
Wherein, the MutiHeadAttention is the attention characteristic value,for the stitching function, the heads are the number of self-attention heads, softmax () is the normalized exponential function,QIn the case of a first spatial matrix,Kin the case of a second spatial matrix,VFor the third space matrix,/>Is a dimension parameter,/>Is a first weight matrix.
In this embodiment, the prediction device uses the spatial matrix of the sample training set as input data of a plurality of self-attention heads of the self-attention module, obtains attention feature values according to a preset self-attention extraction algorithm, and obtains attention features with richer details by performing contextual attention extraction on the spatial matrix so as to perform more accurate model training.
S213: and inputting the attention characteristic value into the prediction module, and obtaining the predicted value of the lubricating oil quantity of the aero-engine according to a preset predicting algorithm of the lubricating oil quantity of the aero-engine.
the prediction algorithm of the lubricating oil quantity of the aero-engine is as follows:
In the method, in the process of the invention,for the predicted value of the lubricating oil quantity of the aero-engine,/>Is a second weight matrix.
in this embodiment, the prediction device inputs the attention characteristic value to the prediction module, and obtains the prediction value of the lubricating oil amount of the aero-engine according to a preset prediction algorithm of the lubricating oil amount of the aero-engine.
S22: obtaining true values of the lubricating oil quantity of the aero-engine corresponding to the sample training sets, and obtaining the loss values of the sub-models according to the predicted values of the lubricating oil quantity of the aero-engine, the true values of the lubricating oil quantity of the aero-engine and a preset loss function, which are output by the sub-models.
The loss function is:
In the method, in the process of the invention,For loss value,/>Is the firstithe true values of the lubricating oil quantity of the aero-engine corresponding to the sample training sets,Is the firstiPredictive value of the lubricating oil quantity of the aero-engine corresponding to each sample training set,/>As a differentiating function for indicating/>And/>Differences between/>In order for the parameters to be regularized,mIs the number of sample training sets.
In this embodiment, the prediction device obtains true values of the amounts of the lubricating oil of the aero-engine corresponding to the plurality of sample training sets, and obtains the loss values of the plurality of sub-models according to the predicted values of the amounts of the lubricating oil of the aero-engine, the true values of the amounts of the lubricating oil of the aero-engine and a preset loss function, which are output by the plurality of sub-models.
s3: according to the loss values of the plurality of sub-models, a plurality of model clusters are constructed by adopting a clustering analysis method, and according to the loss values of the plurality of sub-models in the plurality of model clusters, a plurality of target sub-models are extracted from the plurality of model clusters, and a target aeroengine lubricating oil quantity prediction model is constructed.
Since the loss value of a single sub-model is only one numerical value, the data size is not affected by the data size and the data dimension of service data, and the number of total loss values is only related to the number of sub-models. And the prediction performance of the sub-model with small loss value is necessarily superior to that of the sub-model with larger loss value, so that the loss value has the characteristics of small data scale and capability of well reflecting the performance of each sub-model.
in this embodiment, the prediction device constructs a plurality of model clusters by adopting a cluster analysis method according to the loss values of a plurality of sub-models.
Specifically, the prediction device adopts a DBSCAN clustering model, the DBSCAN clustering model is used as an unsupervised machine learning clustering algorithm, different neural networks can be aggregated into a model cluster according to indexes such as loss values, accuracy values and the like under the condition that the number of clusters is not specified, and the neural networks with similar indexes are conveniently screened according to the indexes to construct an integral model.
The prediction equipment inputs the loss values of a plurality of sub-models into a preset DBSCAN cluster model, obtains distance parameters among the plurality of sub-models, clusters the plurality of sub-models according to the distance parameters among the plurality of sub-models and a preset distance threshold, and constructs a plurality of model clusters.
In an alternative embodiment, the prediction device normalizes the loss values of each sub-model, and then uses the normalized loss values of each sub-model as the input data of the DBSCAN cluster model. The influence of the dimension is eliminated, so that the parameter configuration of the DBSCAN model can be suitable for data of different dimensions.
And the prediction equipment extracts a plurality of target sub-models from the model clusters according to the loss values of the sub-models in the model clusters, and builds a target aeroengine lubricating oil quantity prediction model.
Referring to fig. 4, fig. 4 is a schematic flow chart of step S3 in the method for predicting the lubrication oil amount of an aero-engine according to an embodiment of the application, including step S31, specifically including the following steps:
S31: and carrying out average treatment on the loss values of a plurality of sub-models in the same model cluster to obtain loss average values of the plurality of model clusters, comparing the loss values of the plurality of sub-models in the plurality of model clusters with the loss average values of corresponding model clusters to obtain comparison results of the plurality of sub-models, and extracting a plurality of target sub-models from the plurality of model clusters according to the comparison results.
in this embodiment, the prediction device performs average processing on the loss values of a plurality of sub-models in the same model cluster to obtain a loss average value of the plurality of model clusters, compares the loss values of the plurality of sub-models in the plurality of model clusters with the loss average value of the corresponding model cluster to obtain a comparison result of the plurality of sub-models, extracts a plurality of target sub-models from the plurality of model clusters according to the comparison result, optimizes the performance of the overall model by retaining the sub-models in the model cluster with better performance, and improves the accuracy of prediction of the lubricating oil quantity of the aeroengine.
S4: and responding to a prediction instruction, obtaining time sequence data of flight parameters to be predicted, inputting the time sequence data to be predicted into the target aircraft engine lubricating oil quantity prediction model, obtaining a plurality of aircraft engine lubricating oil quantity predicted values output by the target sub-model, and carrying out average processing on the aircraft engine lubricating oil quantity predicted values output by the target sub-model to obtain an aircraft engine lubricating oil quantity predicted average value as an aircraft engine lubricating oil quantity predicted result of the time sequence data of the flight parameters to be predicted.
The prediction instruction is sent by a user and received by prediction equipment.
The prediction equipment responds to a prediction instruction to obtain time sequence data of flight parameters to be predicted, the time sequence data of the flight to be predicted is input into the target aircraft engine lubricating oil quantity prediction model, a plurality of aircraft engine lubricating oil quantity prediction values output by the target sub-models are obtained, the aircraft engine lubricating oil quantity prediction values output by the target sub-models are subjected to average processing to obtain an aircraft engine lubricating oil quantity prediction average value, the average value is used as an aircraft engine lubricating oil quantity prediction result of the time sequence data of the flight parameters to be predicted, a plurality of sub-models in the aircraft engine lubricating oil quantity prediction model are screened through a cluster analysis method, sub-models in a model cluster with better performance are reserved, a plurality of target sub-models are extracted, the target aircraft engine lubricating oil quantity prediction model is constructed, the performance of the whole model is optimized, the accuracy and the efficiency of the aircraft engine lubricating oil quantity prediction are improved, and therefore an aircraft engine lubricating system is monitored more accurately and effectively.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an apparatus for predicting the amount of lubricating oil of an aero-engine according to an embodiment of the present application, where the apparatus may implement all or a part of the apparatus for predicting the amount of lubricating oil of an aero-engine by software, hardware or a combination of both, and the apparatus 5 includes:
The data acquisition module 51 is configured to acquire a flight parameter time sequence data training set and an aeroengine lubrication oil quantity prediction model to be trained, where the flight parameter time sequence data training set includes a plurality of flight parameter time sequence data, and the aeroengine lubrication oil quantity prediction model includes a plurality of sub-models;
the sub-model training module 52 is configured to construct a plurality of sample training sets corresponding to the sub-models according to the time-series data training sets of the flight parameters, input the plurality of sample training sets into the corresponding sub-models for training, and obtain loss values of the plurality of sub-models;
The sub-model extraction module 53 is configured to construct a plurality of model clusters according to the loss values of a plurality of sub-models by adopting a cluster analysis method, where the model clusters include a plurality of sub-models with the same cluster labels; according to the loss values of a plurality of sub-models in a plurality of model clusters, extracting a plurality of target sub-models from a plurality of model clusters, and constructing a target aero-engine lubricating oil quantity prediction model;
The prediction module 54 is configured to obtain time series data of a flight parameter to be predicted in response to a prediction instruction, input the time series data of the flight parameter to be predicted to the target model of the aircraft engine lubricating oil quantity, obtain a plurality of predicted values of the aircraft engine lubricating oil quantity output by the target sub-model, and perform average processing on the predicted values of the aircraft engine lubricating oil quantity output by the target sub-model to obtain an average predicted value of the aircraft engine lubricating oil quantity as an predicted result of the aircraft engine lubricating oil quantity of the time series data of the flight parameter to be predicted.
In the embodiment, a flight parameter time sequence data training set and an aeroengine lubricating oil quantity prediction model to be trained are obtained through a data obtaining module, wherein the flight parameter time sequence data training set comprises a plurality of flight parameter time sequence data, and the aeroengine lubricating oil quantity prediction model comprises a plurality of sub-models; through a sub-model training module, constructing a plurality of sample training sets corresponding to the sub-models according to the time sequence data training sets of the flight parameters, inputting the plurality of sample training sets into the corresponding sub-models for training, and obtaining loss values of the plurality of sub-models; constructing a plurality of model clusters by a sub-model extraction module according to the loss values of a plurality of sub-models by adopting a cluster analysis method, wherein the model clusters comprise a plurality of sub-models with the same cluster labels; according to the loss values of a plurality of sub-models in a plurality of model clusters, extracting a plurality of target sub-models from a plurality of model clusters, and constructing a target aero-engine lubricating oil quantity prediction model; and responding to a prediction instruction by a prediction module, obtaining time sequence data of flight parameters to be predicted, inputting the time sequence data to be predicted into the target aircraft engine lubricating oil quantity prediction model, obtaining a plurality of aircraft engine lubricating oil quantity predicted values output by the target submodel, and carrying out average processing on the aircraft engine lubricating oil quantity predicted values output by the target submodel to obtain an aircraft engine lubricating oil quantity predicted average value as an aircraft engine lubricating oil quantity predicted result of the time sequence data of the flight parameters to be predicted. Through a cluster analysis method, a plurality of sub-models in the aero-engine lubricating oil quantity prediction model are screened, the sub-models in a better-performing model cluster are reserved, a plurality of target sub-models are extracted, the target aero-engine lubricating oil quantity prediction model is constructed, the performance of the whole model is optimized, the accuracy and the efficiency of aero-engine lubricating oil quantity prediction are improved, and therefore an aero-engine lubricating system is monitored more accurately and effectively.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present application, where the computer device 6 includes: a processor 61, a memory 62, and a computer program 63 stored on the memory 62 and executable on the processor 61; the computer device may store a plurality of instructions adapted to be loaded by the processor 61 and to execute the method steps of fig. 1 to 4, and the specific execution process may refer to the specific description of fig. 1 to 4, which is not repeated here.
Wherein processor 61 may comprise one or more processing cores. The processor 61 performs various functions of the aircraft engine lubricating oil amount prediction device 5 and processes the data by executing or executing instructions, programs, code sets or instruction sets stored in the memory 62 and invoking data in the memory 62 using various interfaces and various parts within the wired connection server, alternatively the processor 61 may be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field-programmable gate array (Field-Programmable Gate Array, FPGA), programmable logic array (Programble Logic Array, PLA). The processor 61 may integrate one or a combination of several of a central processor 61 (Central Processing Unit, CPU), an image processor 61 (Graphics Processing Unit, GPU), and a modem, etc. The CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the touch display screen; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 61 and may be implemented by a single chip.
The Memory 62 may include a random access Memory 62 (Random Access Memory, RAM) or a Read-Only Memory 62 (Read-Only Memory). Optionally, the memory 62 includes a non-transitory computer readable medium (non-transitory computer-readable storage medium). Memory 62 may be used to store instructions, programs, code sets, or instruction sets. The memory 62 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as touch instructions, etc.), instructions for implementing the various method embodiments described above, etc.; the storage data area may store data or the like referred to in the above respective method embodiments. The memory 62 may alternatively be at least one memory device located remotely from the aforementioned processor 61.
the embodiment of the present application further provides a storage medium, where the storage medium may store a plurality of instructions, where the instructions are suitable for being loaded by a processor and executed by the processor, and the specific execution process may refer to the specific description of fig. 1 to 4, and details are not repeated herein.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
in the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other manners. For example, the apparatus/terminal device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical function division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
in addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
the integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc.
the present invention is not limited to the above-described embodiments, but, if various modifications or variations of the present invention are not departing from the spirit and scope of the present invention, the present invention is intended to include such modifications and variations as fall within the scope of the claims and the equivalents thereof.
Claims (8)
1. The method for predicting the lubricating oil quantity of the aero-engine is characterized by comprising the following steps of:
obtaining a flight parameter time sequence data training set and an aeroengine lubricating oil mass prediction model to be trained, wherein the flight parameter time sequence data training set comprises a plurality of flight parameter time sequence data, and the aeroengine lubricating oil mass prediction model comprises a plurality of sub-models;
Constructing a plurality of sample training sets corresponding to the sub-models according to the time sequence data training set of the flight parameters, inputting the plurality of sample training sets into the corresponding sub-models for training, and obtaining loss values of the plurality of sub-models;
Constructing a plurality of model clusters by adopting a clustering analysis method according to the loss values of a plurality of sub-models, extracting a plurality of target sub-models from the plurality of model clusters according to the loss values of a plurality of sub-models in the plurality of model clusters, and constructing a target aeroengine lubricating oil quantity prediction model;
And responding to a prediction instruction, obtaining time sequence data of flight parameters to be predicted, inputting the time sequence data to be predicted into the target aircraft engine lubricating oil quantity prediction model, obtaining a plurality of aircraft engine lubricating oil quantity predicted values output by the target sub-model, and carrying out average processing on the aircraft engine lubricating oil quantity predicted values output by the target sub-model to obtain an aircraft engine lubricating oil quantity predicted average value as an aircraft engine lubricating oil quantity predicted result of the time sequence data of the flight parameters to be predicted.
2. The method for predicting the amount of lubricating oil for an aircraft engine according to claim 1, wherein: the flight parameter time sequence data comprise engine high-pressure rotor rotating speed time sequence data, flight attitude time sequence data, flight height time sequence data and environment temperature time sequence data.
3. the method for predicting the amount of lubricating oil for an aircraft engine according to claim 2, wherein the step of inputting the plurality of sample training sets into the corresponding sub-models to train and obtaining the loss values of the plurality of sub-models includes the steps of:
Inputting a plurality of sample training sets into corresponding sub-models to obtain predicted values of the lubricating oil quantity of the aeroengine corresponding to the plurality of sample training sets output by the plurality of sub-models;
Obtaining true values of the lubricating oil quantity of the aero-engine corresponding to a plurality of sample training sets, and obtaining the loss values of a plurality of sub-models according to the predicted values of the lubricating oil quantity of the aero-engine, the true values of the lubricating oil quantity of the aero-engine and a preset loss function, wherein the loss function is as follows:
In the method, in the process of the invention,For loss value,/>Is the firstitrue value of lubricating oil quantity of aero-engine corresponding to each sample training set,/>Is the firstiPredictive value of the lubricating oil quantity of the aero-engine corresponding to each sample training set,/>As a differentiating function for indicating/>And/>Differences between/>In order for the parameters to be regularized,mIs the number of sample training sets.
4. The method for predicting the amount of lubricating oil for an aircraft engine according to claim 3, wherein: the sub-model comprises a self-attention module and a prediction module, wherein the self-attention module comprises a plurality of self-attention heads;
inputting a plurality of sample training sets into corresponding sub-models to obtain a plurality of aeroengine lubricating oil quantity predicted values corresponding to the sample training sets output by the sub-models, wherein the method comprises the following steps of:
Spatially encoding the sample training set to obtain a plurality of spatial matrixes of the sample training set, wherein the spatial matrixes comprise a first spatial matrix, a second spatial matrix and a third spatial matrix;
respectively taking the space matrix of the sample training set as input data of a plurality of self-attention heads of the self-attention module, and obtaining attention characteristic values according to a preset self-attention extraction algorithm, wherein the attention extraction algorithm is as follows:
Wherein, the MutiHeadAttention is the attention characteristic value,for the stitching function, the heads are the number of self-attention heads, softmax () is the normalized exponential function,QIn the case of a first spatial matrix,Kin the case of a second spatial matrix,VFor the third space matrix,/>Is a dimension parameter,/>Is a first weight matrix;
inputting the attention characteristic value into the prediction module, and obtaining the prediction value of the lubricating oil quantity of the aero-engine according to a preset prediction algorithm of the lubricating oil quantity of the aero-engine, wherein the prediction algorithm of the lubricating oil quantity of the aero-engine is as follows:
In the method, in the process of the invention,for the predicted value of the lubricating oil quantity of the aero-engine,/>Is a second weight matrix.
5. the method for predicting the amount of lubricating oil for an aircraft engine according to claim 1, wherein the step of extracting a plurality of target sub-models from a plurality of model clusters according to the loss values of a plurality of sub-models in a plurality of model clusters comprises the steps of:
And carrying out average treatment on the loss values of a plurality of sub-models in the same model cluster to obtain loss average values of the plurality of model clusters, comparing the loss values of the plurality of sub-models in the plurality of model clusters with the loss average values of corresponding model clusters to obtain comparison results of the plurality of sub-models, and extracting a plurality of target sub-models from the plurality of model clusters according to the comparison results.
6. an aeroengine lubrication oil quantity prediction device, comprising:
The system comprises a data acquisition module, a data prediction module and a data processing module, wherein the data acquisition module is used for acquiring a flight parameter time sequence data training set and an aeroengine lubricating oil quantity prediction model to be trained, the flight parameter time sequence data training set comprises a plurality of flight parameter time sequence data, and the aeroengine lubricating oil quantity prediction model comprises a plurality of sub-models;
The sub-model training module is used for constructing a plurality of sample training sets corresponding to the sub-models according to the time sequence data training sets of the flight parameters, inputting the plurality of sample training sets into the corresponding sub-models for training, and obtaining loss values of the plurality of sub-models;
The sub-model extraction module is used for constructing a plurality of model clusters by adopting a cluster analysis method according to the loss values of the plurality of sub-models, wherein the model clusters comprise a plurality of sub-models with the same cluster labels; according to the loss values of a plurality of sub-models in a plurality of model clusters, extracting a plurality of target sub-models from a plurality of model clusters, and constructing a target aero-engine lubricating oil quantity prediction model;
The prediction module is used for responding to a prediction instruction, obtaining time sequence data of flight parameters to be predicted, inputting the time sequence data to be predicted into the target aircraft engine lubricating oil quantity prediction model, obtaining a plurality of aircraft engine lubricating oil quantity prediction values output by the target submodel, and carrying out average processing on the aircraft engine lubricating oil quantity prediction values output by the target submodel to obtain an aircraft engine lubricating oil quantity prediction average value which is used as an aircraft engine lubricating oil quantity prediction result of the time sequence data of the flight parameters to be predicted.
7. A computer device comprising a processor, a memory and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method for predicting the amount of lubricating oil of an aeroengine as claimed in any one of claims 1 to 5 when the computer program is executed by the processor.
8. a storage medium storing a computer program which, when executed by a processor, implements the steps of the aeroengine lubrication oil quantity prediction method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410194385.3A CN117763976B (en) | 2024-02-22 | 2024-02-22 | Method and device for predicting lubricating oil quantity of aero-engine and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410194385.3A CN117763976B (en) | 2024-02-22 | 2024-02-22 | Method and device for predicting lubricating oil quantity of aero-engine and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117763976A true CN117763976A (en) | 2024-03-26 |
| CN117763976B CN117763976B (en) | 2024-05-14 |
Family
ID=90326022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410194385.3A Active CN117763976B (en) | 2024-02-22 | 2024-02-22 | Method and device for predicting lubricating oil quantity of aero-engine and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117763976B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115546127A (en) * | 2022-09-21 | 2022-12-30 | 厦门大学 | Engine lubricating oil abrasive particle image analysis method and device, medium and equipment |
| CN115753115A (en) * | 2022-11-24 | 2023-03-07 | 国营川西机器厂 | Method for detecting lubricating oil leakage of exhaust device of turbine shaft aircraft engine accessory casing |
| CN115952724A (en) * | 2022-10-21 | 2023-04-11 | 西安交通大学 | Method, system, equipment and medium for predicting residual life of aircraft engine |
| CN116106021A (en) * | 2023-01-16 | 2023-05-12 | 西北工业大学 | Precision improving method for digital twin of aeroengine performance |
| CN116362142A (en) * | 2023-06-01 | 2023-06-30 | 华南师范大学 | Method, device, equipment and storage medium for predicting lubricating oil quantity of aero-engine |
| CN116542389A (en) * | 2023-05-18 | 2023-08-04 | 济南工程职业技术学院 | Truck fuel consumption prediction method, system, equipment and storage medium |
| CN117150882A (en) * | 2023-07-20 | 2023-12-01 | 嬴彻星创智能科技(上海)有限公司 | Engine oil consumption prediction method, system, electronic equipment and storage medium |
-
2024
- 2024-02-22 CN CN202410194385.3A patent/CN117763976B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115546127A (en) * | 2022-09-21 | 2022-12-30 | 厦门大学 | Engine lubricating oil abrasive particle image analysis method and device, medium and equipment |
| CN115952724A (en) * | 2022-10-21 | 2023-04-11 | 西安交通大学 | Method, system, equipment and medium for predicting residual life of aircraft engine |
| CN115753115A (en) * | 2022-11-24 | 2023-03-07 | 国营川西机器厂 | Method for detecting lubricating oil leakage of exhaust device of turbine shaft aircraft engine accessory casing |
| CN116106021A (en) * | 2023-01-16 | 2023-05-12 | 西北工业大学 | Precision improving method for digital twin of aeroengine performance |
| CN116542389A (en) * | 2023-05-18 | 2023-08-04 | 济南工程职业技术学院 | Truck fuel consumption prediction method, system, equipment and storage medium |
| CN116362142A (en) * | 2023-06-01 | 2023-06-30 | 华南师范大学 | Method, device, equipment and storage medium for predicting lubricating oil quantity of aero-engine |
| CN117150882A (en) * | 2023-07-20 | 2023-12-01 | 嬴彻星创智能科技(上海)有限公司 | Engine oil consumption prediction method, system, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117763976B (en) | 2024-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111526119A (en) | Abnormal flow detection method and device, electronic equipment and computer readable medium | |
| CN114338188B (en) | Intelligent cloud detection system for malicious software based on process behavior sequence segmentation | |
| CN115935286A (en) | Abnormal point detection method, device and terminal for railway bearing state monitoring data | |
| CN115587597B (en) | Sentiment analysis method and device of aspect words based on clause-level relational graph | |
| JP2023504103A (en) | MODEL UPDATE SYSTEM, MODEL UPDATE METHOD AND RELATED DEVICE | |
| CN114722091A (en) | Data processing method, data processing device, storage medium and processor | |
| CN116502175A (en) | Method, device and storage medium for diagnosing fault of graph neural network | |
| CN115545331A (en) | Control strategy prediction method and device, equipment and storage medium | |
| CN117270082A (en) | Precipitation prediction system and method based on multi-sensor data fusion | |
| US20180137579A1 (en) | Method for optimizing crop production efficiency and apparatus for the same | |
| CN116451594A (en) | Training method and device of icing prediction model, prediction method and device and electronic equipment | |
| CN116257663A (en) | Abnormality detection and association analysis method and related equipment for unmanned ground vehicle | |
| CN116684878A (en) | 5G information transmission data safety monitoring system | |
| CN117763976B (en) | Method and device for predicting lubricating oil quantity of aero-engine and computer equipment | |
| CN116362142B (en) | Method, device, equipment and storage medium for predicting lubricating oil quantity of aero-engine | |
| CN114020715A (en) | Log data processing method, device, medium and equipment | |
| CN116502155B (en) | Safety supervision system for numerical control electric screw press | |
| CN115409115A (en) | Time sequence clustering abnormal terminal identification method based on user log | |
| CN114598443A (en) | Malicious software detector training method, detector, electronic device and storage medium | |
| CN110650130A (en) | Industrial control intrusion detection method based on multi-classification GoogLeNet-LSTM model | |
| CN116842174A (en) | Agricultural resource database platform building method based on network data | |
| CN117782599A (en) | Rolling bearing fault diagnosis method, device, equipment and medium | |
| CN117785625A (en) | Method, device, equipment and storage medium for predicting server performance | |
| CN116450485B (en) | Detection method and system for application performance interference | |
| CN117372021A (en) | Training method of risk identification model, risk identification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |