A method for identifying the vibration steady-state condition of an aero-engine

Abstract

This invention relates to the field of aero-engine technology and discloses a method for identifying the steady-state vibration conditions of aero-engines. The method determines whether the engine is in a steady state based on the variation pattern of engine speed, and classifies the operating conditions according to the average steady-state speed and the variation pattern of speed between operating conditions. This invention solves the problems of misjudgment and repeated state fluctuations existing in the prior art.

Description

Technical Field

[0001] This invention relates to the field of aero-engine technology, specifically a method for identifying the steady-state vibration conditions of an aero-engine. Background Technology

[0002] As aero-engine structures become increasingly complex and performance continuously improves, vibration faults are becoming more diverse and severe, directly leading to reduced engine reliability. Therefore, conducting aero-engine fault diagnosis is a crucial measure to ensure normal engine operation and improve equipment availability. Research on data mining methods is currently an important direction in engine fault diagnosis and prediction. Historical engine test vibration data contains a wealth of useful information. In-depth mining of this data helps to understand the engine's condition. Studying the differences between historical data and current test data helps improve the accuracy of fault diagnosis and prediction.

[0003] Studies have found that engine vibration states differ significantly between steady-state and transient states at different speeds, and that vibration states do not belong to the same distribution under different speed conditions. Furthermore, for the same speed, the vibration conditions will differ when the engine speed increases from a low speed to that speed condition versus when it decreases from a high speed to that speed condition. Figure 1 This illustrates the two acceleration / deceleration states. Figure 1 The symbols + and - are used to indicate the speed increase and speed decrease conditions, respectively, as described below. Therefore, only after identifying and classifying stable operating conditions from massive amounts of data can effective data extraction and mining be carried out.

[0004] Since the vibration of rotating machinery is closely related to the rotor speed, the speed value and its stability should be used as the basis for classifying operating conditions. However, engine operating conditions are complex and variable, and the speed change pattern is poor. Manually classifying operating conditions and judging states based on massive amounts of data is very time-consuming. Currently, thresholds are often set for the speed difference, mean speed, and variance within a sliding window time period as conditions for judging steady state. However, in some extreme cases, misjudgments and repeated state jumps may still occur, and there is no distinction between acceleration and deceleration states for the speed operating conditions.

[0005] Speed ​​change curves in some difficult-to-determine situations, such as Figures 2 to 4 As shown.

[0006] like Figure 2 , Figure 3 As shown, it is easy to judge the stability when the speed fluctuates, but then it is judged to be unstable again, making it difficult to distinguish between rising and falling operating conditions.

[0007] like Figure 4As shown, small fluctuations in rotational speed sometimes occur near the threshold, affecting only a single instant, but exhibiting large and sustained changes in rotational speed and variance. When the threshold is set too low, the state tends to jump repeatedly; when the threshold is set too high, other types of transition states are difficult to identify. Summary of the Invention

[0008] To overcome the shortcomings of the prior art, the present invention provides a method for identifying the steady-state vibration conditions of aero-engines, solving problems such as misjudgment and repeated fluctuations in the state that exist in the prior art.

[0009] The technical solution adopted by the present invention to solve the above problems is:

[0010] A method for identifying steady-state vibration conditions of an aero-engine is proposed, which determines whether the engine is in a steady state based on the variation law of the engine speed, and classifies the operating conditions according to the average speed of the steady state and the variation law of the speed between operating conditions.

[0011] As a preferred technical solution, the steps include:

[0012] A1 defines the initial rotational speed value and rotational speed acceleration / deceleration status for a single flight.

[0013] A2, read in the rotational speed data and store the rotational speed data in a sliding window, and take a sliding window of length Δt;

[0014] A3, calculate dn and n within the sliding window. pp The value of ; where dn represents the change in rotational speed, n pp Indicates the magnitude of the speed change;

[0015] A4. Perform a steady-state judgment on the data within the sliding window and use the steady-state judgment result as the current state. If the steady-state judgment result is a steady state, proceed to step A51X, and then proceed to step A5. If the steady-state judgment result is a transient state, proceed to step A51Y, and then proceed to step A5. Wherein, A51X is: update the stable segment information, and A51Y is: record the average speed of the stable segment as the reference speed of the transient segment.

[0016] A5, determine if the flight has ended: if yes, end the flight; if no, return to step A2.

[0017] As a preferred technical solution, in step A4, if the previous moment was a transition state, the condition for determining stability is:

[0018] If dn < δn, and n pp If the speed is less than δn / 2 and remains in a steady state continuously within a set time period T, it is considered that the speed has entered the steady state stage.

[0019] Where δn represents the stability threshold and T represents the time interval of vibration analysis.

[0020] As a preferred technical solution, in step A4, if the previous moment was a steady state, the condition for determining stability is:

[0021] If dn < δn*c, and within a set time period T, the transition state is continuously entered for a period of T / 4 time, then the transition state is considered to have been entered.

[0022] Where δn represents the stability threshold, c represents the amplification factor, and T represents the time interval of the vibration analysis.

[0023] As a preferred technical solution, in step A4, when performing steady-state judgment, the steady-state conditions in which the high-voltage rotor speed is lower than the set threshold or the steady-state conditions whose stability duration is less than the set time are eliminated.

[0024] As a preferred technical solution, step A51X includes the following steps:

[0025] A51X1, Calculate the speed characteristics of the steady-state section: take the average speed n of the current steady-state condition as the average speed within time T;

[0026] A51X2, determine whether the current stable segment is in an upward or downward state;

[0027] A51X3 updates the start and end times of the stable segment, the average speed, and information on whether the segment is rising or falling.

[0028] As a preferred technical solution, in step A51X3, the formula for calculating the updated average rotational speed is:

[0029]

[0030] in, is the average speed, n is the speed value at the last point of the sliding window, and N is the cumulative number of speed points under the current operating condition.

[0031] As a preferred technical solution, in A51X2, the rule for determining whether the current stable segment is in an upward or downward state is as follows: the average speed of the current steady-state condition is compared with the reference speed of the transient state. If |current average speed - reference speed| > a set threshold, the difference is positive and it is in an upward state; if the difference is negative, it is in a downward state. Otherwise, it is considered to be the same state as the previous condition. Here, || represents absolute value operation.

[0032] As a preferred technical solution, the value of Δt is 1s.

[0033] As a preferred technical solution, the value of T is 0.3-0.5s.

[0034] Compared with the prior art, the present invention has the following advantages:

[0035] (1) The present invention is compatible with burrs, steps, and situations where the rotational speed increases and then decreases during rotation;

[0036] (2) The present invention can promptly determine stability, avoid repeated fluctuations between steady state and transition state, and distinguish between rising and falling states, thus meeting the statistical requirements for engine vibration data. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of acceleration / deceleration.

[0038] Figure 2 This is one of the speed variation curves;

[0039] Figure 3 This is the second graph showing the change in rotational speed;

[0040] Figure 4 This is the third graph showing the change in rotational speed;

[0041] Figure 5 This is a schematic diagram illustrating the working condition identification using the present invention;

[0042] Figure 6 This is a flowchart of the present invention. Detailed Implementation

[0043] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0044] Example 1

[0045] like Figures 1 to 6 As shown, to address the problems of existing technologies, a steady-state judgment and operating condition classification method based on automatic division of steady-state conditions into rising and falling states is proposed. The main process is as follows: Figure 6 As shown.

[0046] A sliding window approach is used, sliding the relative physical speed data of the high-voltage rotor and the low-voltage rotor from the first point to the last point, with a sliding window length of Δt. This method uses the difference dn between the first and last speed points within the sliding window, and the difference n between the maximum and minimum speed values ​​within the sliding window, to analyze these differences. pp A threshold is set to determine whether the engine speed is stable. The average value of the high-pressure rotor relative to the physical speed during the steady-state speed range is used as the basis for classifying operating conditions. The method first determines whether the engine state is in a steady state according to the stability judgment rules, and then classifies operating conditions and acceleration / deceleration states based on the average speed in the steady state and the speed change pattern between operating conditions.

[0047] The determination of whether the rotational speed is stable is based on whether it was in a steady state at the previous moment, with thresholds set accordingly. The initial state is set by the user, and is generally set to a transition state.

[0048] ①If the previous moment was a transition state, the condition for determining stability is:

[0049] dn < δn, and n pp <δn / 2 where δn is the stability threshold;

[0050] Furthermore, if the rotational speed remains in a steady state for a period of time T (the time interval of vibration analysis), it is considered to have entered the steady-state stage.

[0051] ②The previous moment was a steady state. The condition for determining stability is:

[0052] dn < δn * c, where c is the amplification factor.

[0053] Furthermore, if a transition state is entered for a continuous period of time T for a period of T / 4, it is considered to have entered the transition state.

[0054] In a steady state, the requirements are appropriately relaxed to make it less likely to be classified as a transitional state, thus accommodating occasional speed step-down situations. In a transitional state, a threshold is added for the speed difference within the sliding window to strictly judge speed fluctuations before entering steady state, preventing misjudgments. The standards are appropriately tightened to make it difficult to enter a steady-state condition, preventing repeated fluctuations before reaching steady state. Finally, any steady-state condition with a duration of less than 5 seconds is discarded.

[0055] The stable operating conditions are divided based on the average speed within the operating condition range, with each operating condition defined as 1% of the high voltage relative to the physical speed.

[0056] The acceleration / deceleration status of the operating condition is determined based on the speed difference between two adjacent steady-state operating conditions.

[0057] During the transition state, the average speed of the previous steady state is recorded as the reference speed n for the change in speed. ref When it re-enters a steady state, it is considered to have entered a new vibration stability condition, and the sliding average value of the rotational speed is calculated. The current rotational speed is compared with the reference rotational speed of the previous transient state. If the difference in rotational speed is >2%, the current condition is > the previous steady-state condition, indicating an upward trend; otherwise, it is a downward trend. If the difference in rotational speed is less than 2%, the acceleration and deceleration states of the two conditions are considered to be consistent.

[0058] In addition, for aero engines, research is generally conducted only on operating conditions at idle speed and above, and conditions with high-pressure rotor speeds below 65% are excluded.

[0059] like Figure 5 As shown, this invention can effectively identify the stable section and the transition section at the actual test speed, and correctly identify the acceleration and deceleration states.

[0060] Example 2

[0061] like Figures 1 to 6 As shown, as a further optimization of Embodiment 1, this embodiment also includes the following technical features based on Embodiment 1:

[0062] Step 1: Define the initial speed value and speed acceleration / deceleration state for a single flight. When the data starts from the start, the speed value is set to 0; otherwise, it is set to the initial value as the speed value of the previous stable operating condition, filling the sliding window. The speed state is generally defined as the transient state, but the user can set it to the steady state. The operating condition at this time is set to either an acceleration or deceleration state according to the actual situation. If it starts from the start, it is set to the acceleration state.

[0063] Step 2: Read in the rotational speed data and store it in the sliding window. A sliding window of length t. The rotation speed within the sliding window, each time a new point is entered, is appended to the end of the array, and the first point in the array is removed.

[0064] Step 3: Calculate dn and n within the sliding window. pp The value of the stability determination is determined, and the stability determination result is stored as the current state. The stability determination rules are shown in Example 1.

[0065] The parameters have the following meanings:

[0066] dn: Rotational speed change value, the difference between the speed value at the last point of the sliding window and the rotational speed value at the first point;

[0067] n pp Rotational speed variation amplitude: the difference between the maximum and minimum values ​​within the sliding window;

[0068] n: Rotational speed value;

[0069] Average speed during the steady-state range.

[0070] If the previous moment was in a steady state, and data points of T / 4 appear consecutively within the time interval T, it is considered a transitional state, and the state is considered to have entered a steady state. If the previous moment was in a transitional state, and the entire time interval T was in a steady state, then the state is considered to have entered a steady state. Otherwise, the state remains unchanged.

[0071] Step 4: Store the steady-state determination results in a sliding window into the steady-state determination result flag array stableflag within time period T.

[0072] Step 5: If the state changes from steady state to transition state, record the average speed and start and end times of the previous steady-state condition, and update this information to the previous steady-state condition. The current state is updated to the transition state, and the reference speed is the average speed of the previous steady-state condition.

[0073] Step 6: Upon transitioning from the transient state to the steady state, update the current state to the steady state and begin calculating the average speed under the current steady-state operating condition. Given the average rotational speed over time T, the acceleration / deceleration state of the current steady-state condition is determined based on the reference rotational speed recorded during the transient state.

[0074] The rule for judging the acceleration / deceleration state is as follows: compare the average speed of the current steady-state condition with the average reference speed of the transient state. If |current average speed - reference speed| > 2%, the difference is positive and the state is accelerating; if the difference is negative, the state is decelerating. Otherwise, it is considered to be the same state as the previous condition.

[0075] Step 7: If the state is determined to be steady and no change in state has occurred, update the mean speed of the steady state. The formula for calculating the mean speed is: in is the average speed, n is the speed value at the last point of the sliding window, and N is the cumulative number of speed points under the current operating condition.

[0076] Step 8: If it is determined to be a transition state and no change in state has occurred, no action is taken.

[0077] Step 9: Once the sliding window has experienced all the time points of a flight, the judgment ends. Eliminate steady-state conditions that are less than the idle state or whose stable duration is less than 5 seconds. Extract all steady-state conditions and classify them into different conditions based on the high-pressure relative physical speed.

[0078] Parameter selection:

[0079] The sliding window Δt reference control stability judgment is set to 1s; the δn in the stability judgment rule refers to the engine speed control rate, with the low-pressure rotor speed taken as 0.3% and the high-pressure rotor speed taken as 0.2%. The amplification factor is set to 2 based on experience. Generally, the time resolution of vibration analysis should be lower than the sampling rate of speed acquisition. The stability observation time T is based on the vibration analysis time interval, generally around 0.3-0.5s.

[0080] Threshold for consistent operating speeds: Based on the characteristics of engine vibration data, a speed difference of 100 Hz is considered significant, and the threshold is set at 1% of the maximum speed. Threshold for determining whether there is a clear trend in speed change, distinguishing between rising and falling states, is set at 2%.

[0081] As described above, the present invention can be implemented well.

[0082] All features disclosed in all embodiments of this specification, or steps in all methods or processes implied in the disclosure, may be combined and / or extended or replaced in any way, except for mutually exclusive features and / or steps.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for identifying the steady-state vibration condition of an aero-engine, characterized in that, The engine state is determined based on the variation pattern of the rotational speed, and the operating conditions are divided according to the average rotational speed in the steady state and the variation pattern of rotational speed between operating conditions. The method for identifying the steady-state vibration condition of an aero-engine includes the following steps: A1 defines the initial rotational speed value and rotational speed acceleration / deceleration status for a single flight. A2, read in the rotational speed data and store it in the sliding window, taking a length of... Sliding windows; A3, Calculate the value within the sliding window. , The value of; where, This indicates the change in rotational speed. Indicates the magnitude of the speed change; A4. Perform a steady-state judgment on the data within the sliding window and use the steady-state judgment result as the current state. If the steady-state judgment result is a steady state, proceed to step A51X, and then proceed to step A5. If the steady-state judgment result is a transient state, proceed to step A51Y, and then proceed to step A5. Wherein, A51X is: update the stable segment information, and A51Y is: record the average speed of the stable segment as the reference speed of the transient segment. A5, determine if the flight has ended: if yes, end the flight; if no, return to step A2; In step A4, if the previous moment was a transition state, the condition for determining stability is: like And set a time period If the internal rotational speed remains in a steady state, it is considered to have entered a steady-state phase. in, Indicates the stability threshold. Indicates the time interval for vibration analysis; In step A4, if the previous moment was a steady state, the condition for determining stability is: like And within a set period of time If a process continuously enters a transition state for a period of T / 4, it is considered to have entered a transition state. in, Indicates the stability threshold. Indicates the magnification factor. Indicates the time interval for vibration analysis; Step A51X includes the following steps: A51X1, Calculate the steady-state speed characteristics: using the average speed under the current steady-state operating conditions. Let T be the average rotational speed over time period T; A51X2, determine whether the current stable segment is in an upward or downward state; A51X3 updates the start and end times of the stable segment, the average speed, and information on whether the speed is rising or falling. In step A51X3, the formula for calculating the updated average rotational speed is: ； in, The average rotational speed. This is the final rotation speed value for the sliding window. The cumulative speed points under the current operating conditions; In A51X2, the rule for determining whether the current stable segment is in an upward or downward state is as follows: compare the average speed of the current steady-state condition with the reference speed of the transient state. If |current average speed - reference speed| > the set threshold, the difference is positive and it is in an upward state; if the difference is negative, it is in a downward state. Otherwise, it is considered to be the same state as the previous condition. Here, || represents absolute value operation.

2. The method for identifying the steady-state vibration condition of an aero-engine according to claim 1, characterized in that, In step A4, when performing steady-state judgment, conditions where the high-voltage rotor speed is lower than the set threshold or where the steady-state duration is less than the set time are eliminated from all steady-state conditions.

3. The method for identifying the steady-state vibration condition of an aero-engine according to claim 1, characterized in that, The value is 1s.

4. A method for identifying the steady-state vibration condition of an aero-engine according to any one of claims 1 to 3, characterized in that, The value of T is 0.3-0.5s.

Images