Aero-engine multi-stage degradation demarcation point online detection method

Abstract

The invention belongs to the technical field of aero-engine residual life prediction and health management, and discloses an aero-engine multi-stage degradation demarcation point online detection method, which comprises the following steps: acquiring aero-engine monitoring data, and preprocessing the monitoring data; setting a detection threshold value according to a statistical index of the monitoring data, and detecting the preprocessed monitoring data by using the detection threshold value to obtain an I-type demarcation point of a normal operation stage and a slow degradation stage; and after the existence of the I-type demarcation point is detected, calculating an increment maximum likelihood difference value of the preprocessed monitoring data by adopting an increment maximum likelihood method, and detecting the II-type demarcation point of the slow degradation stage and the rapid degradation stage by utilizing the increment maximum likelihood difference value. According to the method, the demarcation point can be effectively detected, and the calculation amount is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft engine remaining life prediction and health management, and in particular to an aircraft engine multi-stage degradation demarcation point online detection method. Background Art

[0002] In recent years, the aviation industry has developed rapidly. As the heart of the aircraft, the health of aircraft engines has attracted much attention. In order to prevent aircraft engine failure, regular maintenance is generally used to repair and maintain aircraft engines at this stage. However, advanced design and manufacturing technologies are changing with each passing day, and the reliability of aircraft engines themselves is also constantly improving. Although this maintenance method can ensure the safe service of aircraft engines, it is too conservative and causes a huge waste of resources. With the continuous development of the Internet of Things, especially the advancement of data acquisition and communication technologies, a comprehensive aircraft engine sensor network has been established, which provides convenience for real-time collection of aircraft engine health status data.

[0003] At present, the operation and maintenance strategy of aircraft engines based on monitoring data has received extensive attention and research. The core of this type of operation and maintenance strategy is to accurately predict the remaining life of aircraft engines. Generally speaking, the degradation process of aircraft engines is divided into three stages. The first stage is the normal operation stage, when the engine runs smoothly and the monitoring signal fluctuates slightly around a certain constant amplitude; the second stage is the slow degradation stage, when the engine shows a degradation trend and the amplitude of the monitoring signal rises slowly at a certain rate; the third stage is the rapid degradation stage, when the engine has an obvious degradation trend and the amplitude of the monitoring signal rises rapidly at a certain rate until it exceeds the threshold and eventually fails.

[0004] The monitoring data of aircraft engines contains degradation information of different operation stages. Different operation and maintenance strategies and corresponding life prediction models are required for different operation stages. If the selected model does not match the degradation stage, it will cause incalculable losses. For example, when the aircraft engine enters the rapid degradation stage and is about to fail, if the prediction algorithm of the slow degradation stage is still used, it will inevitably lead to a delay in the fault alarm time, resulting in incalculable losses. Therefore, it is very important to accurately detect the multi-stage degradation demarcation point of aircraft engines. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides an online detection method for multi-stage degradation demarcation points of an aircraft engine, in order to improve the matching degree between the life prediction algorithm and the degradation stage, and ultimately improve the accuracy of life prediction.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0007] An online detection method for a multi-stage degradation demarcation point of an aircraft engine comprises the following steps:

[0008] Obtain aircraft engine monitoring data and pre-process the monitoring data;

[0009] The detection threshold is set based on the statistical indicators of the monitoring data, and the detection threshold is used to detect the pre-processed monitoring data to obtain the type I dividing point between the normal operation stage and the slow degradation stage;

[0010] After the existence of the type I cut-off point is detected, the incremental maximum likelihood method is used to calculate the incremental maximum likelihood difference of the preprocessed monitoring data, and the incremental maximum likelihood difference is used to detect the type II cut-off point between the slow degradation stage and the rapid degradation stage.

[0011] Optionally, the monitoring data includes:

[0012] The sampling time and amplitude of the data sampling point.

[0013] Optionally, preprocessing the monitoring data includes:

[0014] Determine whether the monitoring data meets the detection conditions; if so, execute the next step; otherwise, continue to obtain the aircraft engine monitoring data;

[0015] Calculate the sample mean and sample standard deviation of the amplitude of the data sampling points;

[0016] All data sampling points are de-meaned.

[0017] Optionally, the detection conditions include:

[0018] The number of data sampling points is greater than or equal to the number threshold.

[0019] Optionally, performing a mean removal process on all data sampling points includes:

[0020]

[0021] Among them, x n is the amplitude of the data sample point n after removing the mean value, x′ n is the amplitude of the data sampling point, is the sample mean of the amplitude of the data sampling points.

[0022] Optionally, setting the detection threshold based on the statistical indicators of the monitoring data includes:

[0023] The detection threshold is set by the sum of the sample mean of the amplitude of the data sampling point and the sample standard deviation of three times the amplitude.

[0024] Optionally, the pre-processed monitoring data is tested using a detection threshold to obtain a type I demarcation point between the normal operation stage and the slow degradation stage, including:

[0025] Determine whether the amplitudes of three consecutive data sampling points after averaging are all greater than the detection threshold; if so, the current data sampling point is the type I dividing point between the normal operation stage and the slow degradation stage; otherwise, continue to detect the next data sampling point.

[0026] Alternatively, the incremental maximum likelihood difference is calculated as:

[0027]

[0028] in, is the incremental maximum likelihood difference when the data sampling point N is located in the two degradation stages, η1 and σ1 are the degradation rate and diffusion coefficient of the slow degradation stage, η2 and σ2 are the degradation rate and diffusion coefficient of the fast degradation stage, △x N-1 is the amplitude difference between two adjacent data sampling points after removing the mean value, △t N-1 is the interval time between two adjacent data sampling points.

[0029] Optionally, the incremental maximum likelihood difference is used to detect the type II cutoff point between the slow degradation stage and the fast degradation stage, including:

[0030] Determine whether the incremental maximum likelihood differences of three consecutive data sampling points are all less than zero; if so, the current data sampling point is the type II dividing point between the slow degradation stage and the rapid degradation stage; otherwise, continue to detect the next data sampling point.

[0031] The present invention has the following beneficial effects:

[0032] The present invention is aimed at aircraft engines and proposes an online detection method for the multi-stage degradation demarcation point of aircraft engines. Each stage of aircraft engine operation is defined and modeled, and the type I demarcation point is detected by the 3-3 principle, and the type II demarcation point is detected by the incremental maximum likelihood difference method. It can realize the division of the three operating stages of the aircraft engine, and provide an important reference for the selection of prediction models and operation and maintenance strategies. At the same time, the present invention is an online detection method, which only needs to analyze the current sampling point, greatly reducing the amount of calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A schematic diagram of a process for online detection of multi-stage degradation demarcation points of an aircraft engine;

[0034] Figure 2 This is a graph showing the multi-stage demarcation point detection results for aircraft engine simulation degradation data. DETAILED DESCRIPTION

[0035] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0036] like Figure 1 As shown, an embodiment of the present invention provides an online detection method for a multi-stage degradation demarcation point of an aircraft engine, comprising the following steps S1 to S3:

[0037] S1. Obtain aircraft engine monitoring data and pre-process the monitoring data.

[0038] In an optional embodiment of the present invention, the monitoring data acquired in step S1 includes:

[0039] The sampling time and amplitude of the data sampling point.

[0040] Step S1 preprocesses the monitoring data including:

[0041] Determine whether the monitoring data meets the detection conditions; wherein the detection conditions include that the number of data sampling points is greater than or equal to the number threshold;

[0042] If yes, proceed to the next step;

[0043] Otherwise, continue to obtain aircraft engine monitoring data;

[0044] Calculate the sample mean and sample standard deviation of the amplitude of the data sampling points;

[0045] All data sampling points are de-meaned, including:

[0046]

[0047] Among them, x n is the amplitude of the data sample point n after removing the mean value, x′ n is the amplitude of the data sampling point, is the sample mean of the amplitude of the data sampling points.

[0048] Exemplarily, this embodiment first preprocesses the collected data. The current sampling point of online monitoring is denoted as the Nth sampling point, and the corresponding sampling time is t N , the monitoring signal amplitude is x′ NFirst, determine whether the number of sampling points N reaches 50. If N<50, continue monitoring; if N≥50, calculate the sample mean of the amplitude of the first 50 sampling points (that is, the sampling points before the number threshold) And the sample standard deviation S, expressed as:

[0049]

[0050] All historical sampling points and newly entered sampling points are processed to remove the mean, that is, For convenience, the amplitude after de-averaging is referred to as amplitude in subsequent steps.

[0051] This embodiment then establishes a three-stage operation process model of an aircraft engine, assuming that the normal stage follows a normal distribution N(μ,σ 2 ), the slow degradation stage conforms to the Wiener degradation process W(η1,σ1) and the fast degradation stage conforms to the Wiener degradation process W(η2,σ2), expressed as:

[0052]

[0053] Among them, t k represents the time of the kth data sampling point; t I and t II Respectively represent the sampling time of type I and type II demarcation points; △t k-1 =t k -t k-1 Indicates the interval between two adjacent data sampling points; x k Indicates that at t k The amplitude of the aircraft engine monitoring signal at the time; x I and x II They represent the amplitude of the aircraft engine monitoring signal at the type I demarcation point and the type II demarcation point respectively; μ and σ represent the mean and standard deviation of the normal stage respectively; η1 and σ1 represent the degradation rate and diffusion coefficient of the slow degradation stage respectively; η2 and σ2 represent the degradation rate and diffusion coefficient of the fast degradation stage respectively; ω k-1 Represents the transmission error, which obeys the normal distribution N(0,△t k-1 ); Assume that η1, η2, σ1 and σ2 are all known.

[0054] S2. Set the detection threshold based on the statistical indicators of the monitoring data, and use the detection threshold to detect the pre-processed monitoring data to obtain the type I dividing point between the normal operation stage and the slow degradation stage.

[0055] In an optional embodiment of the present invention, step S2 of setting the detection threshold based on the statistical indicators of the monitoring data includes:

[0056] The detection threshold is set by the sum of the sample mean of the amplitude of the data sampling point and the sample standard deviation of three times the amplitude.

[0057] Step S2 uses the detection threshold to detect the pre-processed monitoring data to obtain the type I demarcation point between the normal operation stage and the slow degradation stage, including:

[0058] Determine whether the amplitudes of three consecutive data sampling points after averaging are all greater than the detection threshold; if so, the current data sampling point is the type I dividing point between the normal operation stage and the slow degradation stage; otherwise, continue to detect the next data sampling point.

[0059] For example, this embodiment detects the type I demarcation point based on the 3-3 principle. That is, the amplitude of three consecutive sampling points exceeds It is considered that the aircraft engine begins to enter the degradation stage and the Type I boundary point appears. The specific steps are: if the current sampling point amplitude Compare the signal amplitude x of the first two sampling points N-1 and x N-2 ,like and Then the current sampling point N is the type I demarcation point, t I =t N , otherwise continue with type I cut-off point detection.

[0060] S3. After the existence of the type I dividing point is detected, the incremental maximum likelihood method is used to calculate the incremental maximum likelihood difference of the preprocessed monitoring data, and the incremental maximum likelihood difference is used to detect the type II dividing point between the slow degradation stage and the rapid degradation stage.

[0061] In an optional embodiment of the present invention, step S3 uses the incremental maximum likelihood method to calculate the incremental maximum likelihood difference of the preprocessed monitoring data as follows:

[0062]

[0063] in, is the incremental maximum likelihood difference when the data sampling point N is located in the two degradation stages, η1 and σ1 are the degradation rate and diffusion coefficient of the slow degradation stage, η2 and σ2 are the degradation rate and diffusion coefficient of the fast degradation stage, △x N-1 is the amplitude difference between two adjacent data sampling points after removing the mean value, △t N-1 is the interval time between two adjacent data sampling points.

[0064] Step S3 uses the incremental maximum likelihood difference to detect the type II dividing point between the slow degradation stage and the fast degradation stage, including:

[0065] Determine whether the incremental maximum likelihood differences of three consecutive data sampling points are all less than zero; if so, the current data sampling point is the type II dividing point between the slow degradation stage and the rapid degradation stage; otherwise, continue to detect the next data sampling point.

[0066] Exemplarily, after the type I cutoff point is detected in this embodiment, the type II cutoff point detection is triggered. According to the increment independence and normality of the Wiener process, for the slow degradation stage, x k -x k-1 The probability density function of For the rapid degradation stage, x k -x k-1 The probability density function of

[0067] In offline detection, the total number of sampling points is M, and for multi-stage processes, the type II cutoff point is t II is the mth sampling point, i.e. t II =t m , then the likelihood function L about m is expressed as:

[0068]

[0069] Here, ∝ is the proportional symbol.

[0070] Taking the natural logarithm, the log-likelihood function l is:

[0071]

[0072] Therefore, the estimated value of m is

[0073] In order to ensure real-time performance and reduce the amount of calculation, the online detection algorithm of the present invention further adopts the incremental maximum likelihood method difference method for detection. N-1 It has not entered the rapid degradation stage, so we know that lmax=l(N-1η1,σ1,η2,σ2,x). So for the current time t N , it is only necessary to compare the log-likelihood function size when the amplitude of the sampling point is in two degradation stages. The specific steps are as follows:

[0074] Define the difference between the log-likelihood functions of the current sampling point N in the two degradation stages for:

[0075]

[0076] where △x N-1 =x N -x N-1 , △t N-1=t N -t N-1 .

[0077] In order to eliminate the interference of accidental fluctuations, three consecutive adjacent points are also used for judgment, that is, when When and It is considered that the aircraft engine has entered the rapid degradation stage, the sampling point N is the type II boundary point, and t II =t N Otherwise, it is considered that the aircraft engine is still in the slow degradation stage and has not reached the Type II cutoff point, so the Type II cutoff point detection continues.

[0078] The method of the present invention is verified by using simulation data: the amplitude monitored during the operation of the aircraft engine is simulated using the above-mentioned three-stage operation process model of the aircraft engine. Since the method of the present invention includes a de-averaging operation, μ is taken as 0, and σ is taken as 2 =5,η1=1,η2=4, and Sampling interval △t k-1 =△t=1, the simulation generates 500 sampling points, and the sampling time of the type I demarcation point is t I =200, Type II cutoff point sampling time t II =400.

[0079] The method of the present invention is used to detect the simulation data, and the detection value of the type I cutoff point is t I =204, the cut-off point detection value of type II is t II =404. The amplitude of the simulation data, the real demarcation point and the detection demarcation point are obtained as follows Figure 2 As shown in the figure, it can be seen that the entire operation process is divided into three obvious stages. The method of the present invention can effectively detect the dividing point, thereby proving the effectiveness and accuracy of the method of the present invention.

[0080] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0081] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0083] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0084] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. An online detection method for aero-engine multi-stage degradation demarcation point, characterized in that: The following steps are involved: Obtain aircraft engine monitoring data and pre-process the monitoring data; The detection threshold is set based on the statistical indicators of the monitoring data, and the detection threshold is used to detect the pre-processed monitoring data to obtain the type I dividing point between the normal operation stage and the slow degradation stage; After the existence of the type I cut-off point is detected, the incremental maximum likelihood method is used to calculate the incremental maximum likelihood difference of the preprocessed monitoring data, and the incremental maximum likelihood difference is used to detect the type II cut-off point between the slow degradation stage and the rapid degradation stage.

2. The method for online detection of multi-stage degradation demarcation points of an aircraft engine according to claim 1, characterized in that: Monitoring data include: The sampling time and amplitude of the data sampling point.

3. The method for online detection of multi-stage degradation demarcation points of an aircraft engine according to claim 1, characterized in that: Preprocessing of monitoring data includes: Determine whether the monitoring data meets the detection conditions; if so, execute the next step; otherwise, continue to obtain the aircraft engine monitoring data; Calculate the sample mean and sample standard deviation of the amplitude of the data sampling points; All data sampling points are de-meaned.

4. The method for online detection of aircraft engine multi-stage degradation demarcation points according to claim 3, characterized in that: Testing conditions include: The number of data sampling points is greater than or equal to the number threshold.

5. The method for online detection of aircraft engine multi-stage degradation demarcation points according to claim 3, characterized in that: The de-meaning process for all data sampling points includes: Among them, x n is the amplitude of the data sample point n after removing the mean value, x′ n is the amplitude of the data sampling point, is the sample mean of the amplitude of the data sampling points.

6. The method for online detection of aircraft engine multi-stage degradation demarcation points according to claim 1, characterized in that: The detection thresholds are set based on the statistical indicators of the monitoring data, including: The detection threshold is set by the sum of the sample mean of the amplitude of the data sampling point and the sample standard deviation of three times the amplitude.

7. The method for online detection of aircraft engine multi-stage degradation demarcation points according to any one of claims 1 to 6, characterized in that: The detection threshold is used to detect the preprocessed monitoring data to obtain the type I dividing point between the normal operation stage and the slow degradation stage, including: Determine whether the amplitudes of three consecutive data sampling points after averaging are all greater than the detection threshold; if so, the current data sampling point is the type I dividing point between the normal operation stage and the slow degradation stage; otherwise, continue to detect the next data sampling point.

8. The method for online detection of aircraft engine multi-stage degradation demarcation points according to claim 1, characterized in that: The incremental maximum likelihood difference is calculated as: in, is the incremental maximum likelihood difference when the data sampling point N is located in the two degradation stages, η1 and σ1 are the degradation rate and diffusion coefficient of the slow degradation stage, η2 and σ2 are the degradation rate and diffusion coefficient of the fast degradation stage, Δx N-1 is the amplitude difference between two adjacent data sampling points after removing the mean value, Δt N-1 is the interval time between two adjacent data sampling points.

9. The method for online detection of aircraft engine multi-stage degradation demarcation points according to claim 1 or 8, characterized in that: The type II cutoff points for detecting the slow degradation stage and the fast degradation stage using the incremental maximum likelihood difference include: Determine whether the incremental maximum likelihood differences of three consecutive data sampling points are all less than zero; if so, the current data sampling point is the type II dividing point between the slow degradation stage and the rapid degradation stage; otherwise, continue to detect the next data sampling point.

Images