Engine fault detection and diagnosis method

Abstract

The invention relates to the technical field of engine monitoring, in particular to an engine fault detection and diagnosis method, which comprises the following steps of: acquiring engine data, vibration data and cooling data; judging the vibration state of the engine; the engine cooling state is judged; the thermal expansion state of the turbine blade is judged; performing conjoint analysis on the engine cooling state and the thermal expansion state of the turbine blade to obtain an engine state index, and dividing engine fault grades according to the engine state index and the engine vibration state; analyzing the engine speed fluctuation state, and optimizing the division process of engine fault grades according to the engine speed fluctuation state; and performing data integration on the transmission shaft rigidity and the lubricating oil pressure to construct a rotating speed influence factor, and optimizing the analysis process of the engine rotating speed fluctuation state according to the rotating speed influence factor. The fault monitoring efficiency of the engine is effectively improved.

Description

Technical Field

[0001] The invention relates to the technical field of engine monitoring, and in particular to an engine fault detection and diagnosis method. Background Art

[0002] As the core power equipment of aircraft, the working status of aircraft engines is directly related to flight safety and operational efficiency. With the continuous advancement of aviation technology, the complexity of engine design and manufacturing has gradually increased. During the operation of the engine, various types of faults may occur, such as vibration, abnormal temperature, insufficient cooling, etc. These faults will not only lead to the decline of engine performance, but may also cause more serious safety accidents. Therefore, timely and accurate detection and diagnosis of engine faults are of great significance to ensure flight safety and reduce maintenance costs.

[0003] Chinese Patent Publication No. CN109975026A discloses an engine fault monitoring method. When the engine starts, the method can obtain the initial water temperature of the engine, determine the current working condition of the engine, and determine the target water temperature rise rate in the water temperature rise rate. In the present invention, multiple parameters of the engine are obtained, the predicted water temperature value is calculated, and the actual water temperature value is obtained. And the predicted water temperature value and the actual water temperature value are compared to determine whether the engine has a fault. The present invention monitors the engine for faults by comparing specific numerical values, and the monitoring results are more accurate. In addition, the application also provides an engine fault monitoring device to ensure the application and implementation of the above method in practice; it can be seen that when monitoring engine faults, this scheme only analyzes temperature and environmental factors, and does not realize the joint analysis of multi-dimensional data, and there is a problem of low efficiency in engine fault monitoring. Summary of the invention

[0004] An object of the present invention is to provide an engine fault detection and diagnosis method to solve at least one of the problems existing in the prior art.

[0005] To achieve the above object, the present invention adopts the following technical solutions: An engine fault detection and diagnosis method, comprising: A thermal expansion correlation model is constructed based on engine temperature data and thermal expansion of turbine blades to determine the thermal expansion state of turbine blades; Perform joint analysis on the engine cooling state and the thermal expansion state of the turbine blades to obtain the engine state index, and classify the engine fault level according to the engine state index and the engine vibration state; Perform fluctuation analysis on the engine speed collected during the management period to obtain the engine speed fluctuation state, and optimize the engine fault level classification process according to the engine speed fluctuation state; The data of drive shaft stiffness and lubricating oil pressure are integrated to construct the speed influencing factor and optimize the analysis process of engine speed fluctuation state.

[0006] Optionally, collecting engine data and cooling data; Preprocessing the engine vibration signal collected during the monitoring period, and performing fast Fourier transform on the preprocessed engine vibration signal to extract the characteristic frequency, and judging the engine vibration state according to the characteristic frequency; The standard deviation of the coolant flow collected during the management cycle is calculated to obtain the flow fluctuation rate and determine the engine cooling status.

[0007] Optionally, the engine temperature t0 is compared and analyzed with a preset temperature t1 to determine the engine temperature state; When judging the thermal expansion state of the turbine blade, if ΔL>L0×[1+β×(T0 / T)], the thermal expansion state of the turbine blade is judged to be an abnormal state, otherwise, the thermal expansion state of the turbine blade is judged to be a normal state; Among them, △L is the thermal expansion of the turbine blade, L0 is the reference length of the blade at 25°C, β is the preset adjustment coefficient, T0 is the duration of the abnormal engine temperature state during the management cycle, and T is the duration of the management cycle.

[0008] Optionally, a cooling abnormality index is constructed according to the judgment result of the engine cooling state within the management cycle. When the engine cooling state is normal, the cooling abnormality index is set to L1, and L1=0; when the engine cooling state is abnormal, the cooling abnormality index is set to L2, and L2=(Qp-u1) / u1×(Qb-u2) / (Qb+u2); Qb is the flow fluctuation rate, Qp is the average coolant flow Qp, u1 is the first preset flow, and u2 is the second preset flow; According to the judgment result of the thermal expansion state of the turbine blade in the management cycle, the thermal expansion abnormality index is constructed. When the thermal expansion state of the turbine blade is normal, the thermal expansion abnormality index is set to W1, and W1 is set to 0; when the thermal expansion state of the turbine blade is abnormal, the thermal expansion abnormality index is set to W2, and the setting is: W2=lg{2×{△L-L0×[1+β×(T0 / T)]}+1} / lg3; The engine state index FY is constructed according to the cooling abnormality index and the thermal expansion abnormality index, and FY=E1×cooling abnormality index+E2×thermal expansion abnormality index, where E1 is the cooling weight and E2 is the thermal expansion weight.

[0009] Optionally, the process of classifying the engine fault level according to the engine state index FY and the engine vibration state is as follows: If h1×j0 / J+h2×FY≤r1, the engine fault level is classified as level one; If r1 If h1×j0 / J+h2×FY>r2, the engine fault level is divided into three levels; Among them, h1 is the vibration weight, h2 is the state weight, r1 is the first preset state index, r2 is the second preset state index, j0 is the number of monitoring cycles in which the engine vibration state is abnormal within the management cycle, and J is the number of monitoring cycles within the management cycle.

[0010] Optionally, when performing a fluctuation analysis on the engine speed, the fluctuation coefficient is set to ZB, and the fluctuation coefficient ZB and the preset fluctuation threshold ZB1 are compared and analyzed to determine the engine speed fluctuation state, and when the engine speed fluctuation state is abnormal, the engine fault level classification process is optimized, and the optimized first preset state index is set to r1', so that the first preset state index increases with the abnormality of the engine speed fluctuation state.

[0011] Optionally, the process of integrating the data of the transmission shaft stiffness and the lubricating oil pressure is as follows: Compare and analyze the transmission shaft stiffness g0 with the preset stiffness g1 to construct a stiffness index, and compare and analyze the lubricating oil pressure y0 with the preset pressure y1 to construct a pressure index; The speed influence factor SY is constructed according to the stiffness index and the pressure index, and the analysis process of the engine speed fluctuation state is optimized according to the speed influence factor. The optimized preset fluctuation threshold is set to ZB2, so that the preset fluctuation threshold decreases with the increase of the speed influence factor.

[0012] Optionally, when extracting characteristic frequencies from the preprocessed engine vibration signal, the fundamental frequency is set to f1, and f1=R / 60 is set. Harmonic frequencies f2 and f3 are identified based on the fundamental frequency f1, and the amplitude of f1 is recorded as Af1, the amplitude of f2 is recorded as Af2, and the amplitude of f3 is recorded as Af3.

[0013] Optionally, the process of judging the engine vibration state according to the characteristic frequency is as follows: If Af2 / Af1>α or Af3 / Af1>α, the engine vibration state in the current monitoring period is determined to be abnormal; otherwise, the engine vibration state in the current monitoring period is determined to be normal, and α is a preset abnormal proportional coefficient.

[0014] Optionally, the flow fluctuation rate Qb of the coolant flow is calculated, and the flow fluctuation rate Qb and the average coolant flow rate Qp are jointly analyzed to determine the engine cooling state, wherein: ​If Qp<u1 or Qb>u2, the engine cooling state of the current management cycle is determined to be an abnormal state, otherwise, the engine cooling state of the current management cycle is determined to be a normal state; Among them, u1 is the first preset flow rate, and u2 is the second preset flow rate.

[0015] The beneficial effects of the present invention are as follows: by comprehensively collecting various data such as engine vibration, temperature, coolant flow, etc., and using various pre-processing methods such as high-pass filtering and fast Fourier transform, the accuracy of characteristic frequency extraction is significantly improved. Through multi-dimensional analysis of vibration state, cooling state and thermal expansion state, this method can more effectively identify typical faults such as imbalance and shaft misalignment, and provide real-time feedback to operation and maintenance personnel. In addition, this method also avoids the dependence on fixed thresholds in traditional methods by dynamically adjusting thresholds and optimizing fault level classification, thereby improving the adaptability under dynamic working conditions. At the same time, a speed influencing factor is constructed, and the comprehensiveness and accuracy of fault diagnosis are further enhanced by integrating important parameters such as transmission shaft stiffness and lubricating oil pressure. The use of this series of technical means not only makes engine fault detection more timely and accurate, but also improves the early warning capability of potential faults, thereby effectively reducing safety hazards and economic losses caused by faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 Schematic diagram of the flow chart of the engine fault detection and diagnosis method of this embodiment.

[0018] Figure 2 Schematic diagram of the flow of the engine vibration state judgment method of this embodiment.

[0019] Figure 3 Schematic diagram of the flow of the engine cooling state determination method of this embodiment.

[0020] Figure 4 Schematic diagram of the flow of the method for constructing the rotation speed influencing factor in this embodiment. DETAILED DESCRIPTION

[0021] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and accompanying drawings. Similar components in the accompanying drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the content specifically described below is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present invention.

[0022] It should be noted that, although the terms first, second, third, etc. may be used to describe in the embodiments of the present application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, without departing from the scope of the embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first.

[0023] See also Figure 1 As shown, it is a flow chart of the engine fault detection and diagnosis method of this embodiment, and the engine fault detection and diagnosis method includes: Step S101, collecting engine data and cooling data, wherein the engine data includes thermal expansion of turbine blades, transmission shaft stiffness and engine speed, engine temperature, engine vibration signal and lubricating oil pressure, and the cooling data is coolant flow rate; in this embodiment, no specific limitation is made on the collection method of engine data and cooling data, and those skilled in the art can freely set it, as long as the collection requirements of engine data and cooling data are met, wherein a temperature sensor can be used to collect engine temperature, a speed sensor can be used to monitor engine speed, a displacement sensor can be used to measure thermal expansion of turbine blades, a pressure sensor can be used to obtain lubricating oil pressure, a stiffness sensor can be used to detect transmission shaft stiffness, and a three-axis acceleration sensor can be used to collect engine vibration signals, and an electromagnetic flowmeter can be used to collect lubricating oil pressure.

[0024] Step S102, preprocessing the engine vibration signal collected during the monitoring period, and performing fast Fourier transform on the preprocessed engine vibration signal to extract the characteristic frequency, and judge the engine vibration state according to the characteristic frequency; the step S102 effectively eliminates the sensor zero drift through high-pass filtering, retains the key high-frequency vibration components, improves the signal baseline stability, and improves the spectrum clarity of the subsequent FFT analysis, and reduces the impact of sudden interference on the extraction of characteristic frequencies based on statistical outlier detection and replacement strategies.

[0025] It can be understood that the present embodiment does not specifically limit the setting of the monitoring period, and those skilled in the art can freely set it as long as the setting requirements of the monitoring period are met, wherein the monitoring period can be set to 1 second.

[0026] See also Figure 2 As shown, it is a flow chart of the engine vibration state judgment method of this embodiment, including: Step S201, pre-processing the engine vibration signal collected during the monitoring period.

[0027] Specifically, in step S201, when preprocessing the original vibration signal collected during the monitoring period, a high-pass filter is used to eliminate the zero drift of the sensor, and when the amplitude of the sampling point exceeds 3 times the standard deviation, it is replaced with the mean of the previous and next points.

[0028] It is understandable that the present embodiment does not specifically limit the method of preprocessing the engine vibration signal, and those skilled in the art can freely set it. Among them, a moving average filter (for example, a simple moving average with a window size of 5 or 7) can also be used to smooth the original vibration signal to reduce the impact of random noise on subsequent analysis, and use the absolute median deviation to detect outliers. For the detected outliers, they can be replaced by the weighted average value in the window where they are located (such as considering 3 sampling points before and after), and the weight can be set according to the distance from the current position: the closer to the current position, the greater the weight.

[0029] Please continue reading Figure 2 As shown, the engine vibration state judgment method also includes: Step S202: Perform fast Fourier transform on the preprocessed engine vibration signal to extract characteristic frequencies.

[0030] Specifically, in step S202, when extracting characteristic frequencies from the preprocessed engine vibration signal, the fundamental frequency is set to f1, and f1=R / 60 is set. Harmonic frequencies f2 and f3 are identified according to the fundamental frequency f1, and the amplitude of f1 is recorded as Af1, the amplitude of f2 is recorded as Af2, and the amplitude of f3 is recorded as Af3.

[0031] Please continue reading Figure 2 As shown, the engine vibration state judgment method also includes: Step S203, determining the vibration state of the engine according to the characteristic frequency.

[0032] Specifically, the process of determining the engine vibration state according to the characteristic frequency in step S203 is as follows: If Af2 / Af1>α or Af3 / Af1>α, the engine vibration state in the current monitoring period is determined to be abnormal; otherwise, the engine vibration state in the current monitoring period is determined to be normal, and α is the preset abnormal proportional coefficient. By setting the preset abnormal proportional coefficient, typical faults such as imbalance and shaft misalignment can be effectively identified, the abnormal vibration diagnosis response rate can be improved, and high-frequency monitoring scenarios can be supported.

[0033] It can be understood that the present embodiment does not specifically limit the setting of the preset abnormality proportional coefficient, and those skilled in the art can freely set it as long as the setting requirements of the preset abnormality proportional coefficient are met, wherein the best value of α is 0.3.

[0034] It can be understood that in this embodiment, the implementation process of "identifying harmonic frequencies f2 and f3 according to fundamental frequency f1, recording the amplitude of f1 as Af1, recording the amplitude of f2 as Af2, and recording the amplitude of f3 as Af3" is not specifically limited, and technical personnel in this field can freely set it, wherein a 1024-point FFT can be performed on the preprocessed vibration signal to obtain a spectrum diagram (resolution ≈9.8Hz), and the peak near f1 (±5Hz window) is searched in the spectrum diagram, and the amplitude Af1 is recorded. The peaks are also searched near 2f1 and 3f1, and Af2 and Af3 are recorded respectively. The fft function in MATLAB can be used to perform fast Fourier transform (FFT) on the vibration signal, and the peaks in the spectrum diagram can be searched by the findpeaks function to extract the fundamental frequency, harmonic frequencies and their amplitudes.

[0035] See also Figure 1 As shown, the engine fault detection and diagnosis method also includes: Step S103, calculating the standard deviation of the coolant flow collected during the management period to obtain the flow fluctuation rate and determine the engine cooling state.

[0036] See also Figure 3 As shown, it is a flow chart of the engine cooling state judgment method of this embodiment, including: Step S301, calculating the standard deviation of the coolant flow collected during the management period to obtain the flow fluctuation rate.

[0037] Specifically, in step S301, the flow fluctuation rate of the coolant flow is set to Qb, and the setting is: , , where qi is the i-th coolant flow data collected during the management period, I is the number of coolant flow data collected during the management period, and Qp is the average coolant flow rate.

[0038] Please continue reading Figure 3 As shown, the engine cooling state judgment method includes: Step S302, determining the engine cooling state according to the flow fluctuation rate.

[0039] Specifically, the step S302 performs a joint analysis on the flow fluctuation rate Qb and the average coolant flow rate Qp to determine the engine cooling state, wherein: If Qp<u1 or Qb>u2, the engine cooling state of the current management cycle is determined to be an abnormal state, otherwise, the engine cooling state of the current management cycle is determined to be a normal state; Among them, u1 is the first preset flow rate, and u2 is the second preset flow rate. The dual threshold joint analysis is used to distinguish between insufficient coolant flow and unstable flow to reduce the misjudgment rate. At the same time, the standard deviation calculation dynamically adapts to different working conditions, avoiding the limitations of fixed thresholds in variable load scenarios, and improving the cooling system fault identification speed.

[0040] It can be understood that, in this embodiment, the setting of the first preset flow rate and the second preset flow rate is not specifically limited, and those skilled in the art can set them freely as long as the setting requirements of the first preset flow rate and the second preset flow rate are met, among which the optimal value of u1 is 15L / min, and the optimal value of u2 is 0.5L / min.

[0041] See also Figure 1 As shown, the engine fault detection and diagnosis method also includes: Step S104: constructing a thermal expansion correlation model based on the engine temperature data and the thermal expansion amount of the turbine blades to determine the thermal expansion state of the turbine blades.

[0042] Specifically, the step S104 compares and analyzes the engine temperature t0 and the preset temperature t1. If t0≤t1, the engine temperature state is determined to be normal. Otherwise, the engine temperature state is determined to be abnormal. The process of determining the thermal expansion state of the turbine blades in step S104 is as follows: If △L>L0×[1+β×(T0 / T)], the thermal expansion state of the turbine blade is determined to be an abnormal state, otherwise, the thermal expansion state of the turbine blade is determined to be a normal state; Among them, △L is the thermal expansion of the turbine blade, L0 is the reference length of the blade at 25°C, β is the preset adjustment coefficient, T0 is the duration of the abnormal engine temperature state during the management cycle, and T is the duration of the management cycle; by constructing a thermal expansion correlation model to adapt to different temperature conditions and reduce the false alarm rate in high temperature environment, and at the same time combining comprehensive analysis of temperature abnormality duration, blade deformation caused by thermal fatigue can be more accurately identified.

[0043] Specifically, the reference length of the leaf at 25°C can be obtained interactively.

[0044] It is understandable that the present embodiment does not specifically limit the setting of the management period, and those skilled in the art can freely set it as long as the setting requirements of the management period are met, wherein the management period can be set to 5 minutes, 10 minutes, etc.

[0045] It can be understood that the present embodiment does not specifically limit the setting of the preset adjustment coefficient, and those skilled in the art can freely set it as long as the setting requirements of the preset adjustment coefficient are met, wherein the optimal value of β is 0.03.

[0046] See also Figure 1 As shown, the engine fault detection and diagnosis method also includes: Step S105 , jointly analyzing the engine cooling state and the thermal expansion state of the turbine blades to obtain an engine state index, and classifying the engine fault level according to the engine state index and the engine vibration state.

[0047] Specifically, the step S105 constructs a cooling abnormality index according to the judgment result of the engine cooling state within the management cycle. When the engine cooling state is normal, the cooling abnormality index is set to L1, and L1=0; when the engine cooling state is abnormal, the cooling abnormality index is set to L2, and L2=(Qp-u1) / u1×(Qb-u2) / (Qb+u2); The step S105 constructs a thermal expansion abnormality index according to the judgment result of the thermal expansion state of the turbine blade within the management period. When the thermal expansion state of the turbine blade is in a normal state, the thermal expansion abnormality index is set to W1, and W1 is set to 0; when the thermal expansion state of the turbine blade is in an abnormal state, the thermal expansion abnormality index is set to W2, and W2 is set to lg{2×{△L-L0×[1+β×(T0 / T)]}+1} / lg3; The step S105 constructs an engine state index FY according to the cooling abnormality index and the thermal expansion abnormality index, and sets FY=E1×cooling abnormality index+E2×thermal expansion abnormality index, where E1 is the cooling weight, E2 is the thermal expansion weight, and E1+E2=1; The process of step S105 of classifying the engine fault level according to the engine state index FY and the engine vibration state is as follows: If h1×j0 / J+h2×FY≤r1, the engine fault level is classified as level one; If r1 If h1×j0 / J+h2×FY>r2, the engine fault level is divided into three levels; Wherein, h1 is the vibration weight, h2 is the state weight, h1+h2=1, r1 is the first preset state index, r2 is the second preset state index, r1<r2, j0 is the number of monitoring cycles in which the engine vibration state is abnormal within the management cycle, and J is the number of monitoring cycles within the management cycle; ​In step S105, when the engine fault level is level one, no warning is given to the user; when the engine fault level is level two, a mild fault warning is given to the user; when the engine fault level is level three, a severe fault warning is given to the user; multi-parameter weighted fusion is used to avoid the limitation of a single indicator and improve the accuracy of comprehensive diagnosis; at the same time, the gradient response of the fault level division makes operation and maintenance decisions more accurate, thereby ultimately improving the monitoring efficiency of engine faults.

[0048] It can be understood that, in this embodiment, there is no specific limitation on the setting of each weight and each preset state index, and those skilled in the art can set them freely as long as the setting requirements of each weight and each preset state index are met. Among them, the best value of h1 is 0.7, the best value of h2 is 0.3, the best value of r1 is 0.3, the best value of r2 is 0.5, the best value of E1 is 0.4, and the best value of E2 is 0.6.

[0049] See also Figure 1 As shown, the engine fault detection and diagnosis method also includes: Step S106, performing a fluctuation analysis on the engine speed collected during the management period to obtain the engine speed fluctuation state, and optimizing the engine fault level classification process according to the engine speed fluctuation state.

[0050] Specifically, in step S106, when performing fluctuation analysis on the engine speed, the fluctuation coefficient is set to ZB, and the setting is: , , dz is the zth engine speed data collected in the management cycle, Z is the number of engine speed data collected in the management cycle, and dp is the average value of the engine speed data collected in the management cycle; The step S106 compares and analyzes the fluctuation coefficient ZB and the preset fluctuation threshold ZB1 to analyze the engine speed fluctuation state. If ZB≤ZB1, the engine speed fluctuation state is determined to be a normal state. Otherwise, the engine speed fluctuation state is determined to be an abnormal state. The step S106 optimizes the engine fault level classification process when the engine speed fluctuation state is abnormal, and sets the optimized first preset state index to r1', setting r1'=r1×{1+exp[3×(ZB-ZB1)-3]}; the speed fluctuation degree is quantified by standard deviation calculation to effectively distinguish normal fluctuations from abnormal fluctuations. When the speed fluctuation state is abnormal, the first preset state index is dynamically adjusted with the fluctuation degree to ensure that the fault level classification is more in line with the actual working conditions and significantly improve the sensitivity to abnormal fluctuations.

[0051] It can be understood that the present embodiment does not specifically limit the setting of the preset fluctuation threshold, and those skilled in the art can freely set it as long as the setting requirements of the preset fluctuation threshold are met, wherein the optimal value of ZB1 is 0.15.

[0052] See also Figure 1 As shown, the engine fault detection and diagnosis method also includes: Step S107, integrating the data of the transmission shaft stiffness and the lubricating oil pressure to construct a speed influencing factor and optimize the analysis process of the engine speed fluctuation state.

[0053] See also Figure 4 As shown, it is a flow chart of the method for constructing the speed influence factor of this embodiment, including: Step S401 , integrating data of the transmission shaft stiffness and lubricating oil pressure to construct a rotation speed influencing factor.

[0054] Specifically, the process of integrating the data of the transmission shaft stiffness and the lubricating oil pressure in step S401 is as follows: Compare and analyze the transmission shaft stiffness g0 and the preset stiffness g1 to construct a stiffness index. If g0 < g1, the stiffness index is set to GD1, and GD1 = exp[-(g1-g0) / g1]. Otherwise, the stiffness index is set to GD2, and GD2 = 0. Compare and analyze the lubricating oil pressure y0 and the preset pressure y1 to construct a pressure index. If y0 < y1, the pressure index is set to YL1 = ln [(y1-y0) / y1+1]. Otherwise, the pressure index is set to YL2, and YL2 = 0. The speed influencing factor SY is constructed according to the stiffness index and the pressure index, and SY=stiffness index×pressure index. By quantifying the influence of the transmission shaft stiffness and lubricating oil pressure on the speed fluctuation state analysis, the comprehensiveness and accuracy of fault diagnosis are significantly improved.

[0055] It can be understood that the present embodiment does not specifically limit the setting of the preset stiffness and the preset pressure, and those skilled in the art can freely set them as long as the setting requirements of the preset stiffness and the preset pressure are met, wherein the optimal value of g1 is 6000 N·m / rad, and the optimal value of y1 is 345 kPa.

[0056] Please continue reading Figure 4 As shown, the method for constructing the rotation speed influencing factor also includes: Step S402, optimizing the analysis process of the engine speed fluctuation state according to the speed influencing factor.

[0057] Specifically, the step S402 optimizes the analysis process of the engine speed fluctuation state according to the speed influencing factor, sets the optimized preset fluctuation threshold to ZB2, and sets ZB2=ZB1×(1-η×SY), where η is the preset correction coefficient; by quantifying the comprehensive influence of stiffness and pressure on speed fluctuation, dynamically adjusting the preset fluctuation threshold, and improving the sensitivity to abnormal fluctuations, thereby improving the accuracy of engine fault monitoring.

[0058] It can be understood that the present embodiment does not specifically limit the setting of the preset correction coefficient, and those skilled in the art can freely set it as long as the setting requirements of the preset correction coefficient are met, wherein the optimal value of η is 0.2.

[0059] Specifically, the method described in this embodiment is applied to the detection and diagnosis of aircraft engine vibration faults, and the detection accuracy is improved by introducing analysis parameters such as thermal expansion state and coolant flow rate.

[0060] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not limitations on the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the embodiments here. All obvious changes or modifications derived from the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A method for detecting and diagnosing engine faults, characterized in that: include: A thermal expansion correlation model is constructed based on engine temperature data and thermal expansion of turbine blades to determine the thermal expansion state of turbine blades; Perform joint analysis on the engine cooling state and the thermal expansion state of the turbine blades to obtain the engine state index, and classify the engine fault level according to the engine state index and the engine vibration state; Perform fluctuation analysis on the engine speed collected during the management period to obtain the engine speed fluctuation state, and optimize the engine fault level classification process according to the engine speed fluctuation state; The data of drive shaft stiffness and lubricating oil pressure are integrated to construct the speed influencing factor and optimize the analysis process of engine speed fluctuation state.

2. The engine fault detection and diagnosis method according to claim 1, characterized in that: Also includes: Collect engine data and cooling data; Preprocessing the engine vibration signal collected during the monitoring period, and performing fast Fourier transform on the preprocessed engine vibration signal to extract the characteristic frequency, and judging the engine vibration state according to the characteristic frequency; The standard deviation of the coolant flow collected during the management cycle is calculated to obtain the flow fluctuation rate and determine the engine cooling status.

3. The engine fault detection and diagnosis method according to claim 2, characterized in that: Compare and analyze the engine temperature t0 and the preset temperature t1 to determine the engine temperature state; When judging the thermal expansion state of the turbine blade, if ΔL>L0×[1+β×(T0 / T)], the thermal expansion state of the turbine blade is judged to be an abnormal state, otherwise, the thermal expansion state of the turbine blade is judged to be a normal state; Among them, △L is the thermal expansion of the turbine blade, L0 is the reference length of the blade at 25°C, β is the preset adjustment coefficient, T0 is the duration of the abnormal engine temperature state during the management cycle, and T is the duration of the management cycle.

4. The engine fault detection and diagnosis method according to claim 3, characterized in that: A cooling abnormality index is constructed according to the judgment result of the engine cooling state within the management cycle. When the engine cooling state is normal, the cooling abnormality index is set to L1, and L1=0; when the engine cooling state is abnormal, the cooling abnormality index is set to L2, and L2=(Qp-u1) / u1×(Qb-u2) / (Qb+u2); Qb is the flow fluctuation rate, Qp is the average coolant flow Qp, u1 is the first preset flow, and u2 is the second preset flow; According to the judgment result of the thermal expansion state of the turbine blade in the management cycle, the thermal expansion abnormality index is constructed. When the thermal expansion state of the turbine blade is normal, the thermal expansion abnormality index is set to W1, and W1 is set to 0; when the thermal expansion state of the turbine blade is abnormal, the thermal expansion abnormality index is set to W2, and the setting is: W2=lg{2×{△L-L0×[1+β×(T0 / T)]}+1} / lg3; The engine state index FY is constructed according to the cooling abnormality index and the thermal expansion abnormality index, and FY=E1×cooling abnormality index+E2×thermal expansion abnormality index, where E1 is the cooling weight and E2 is the thermal expansion weight.

5. The engine fault detection and diagnosis method according to claim 4, characterized in that: The process of classifying engine fault levels according to the engine status index FY and the engine vibration status is as follows: If h1×j0 / J+h2×FY≤r1, the engine fault level is classified as level one; If r1<h1×j0 / J+h2×FY≤r2, the engine fault level is classified into level 2; If h1×j0 / J+h2×FY>r2, the engine fault level is divided into three levels; Among them, h1 is the vibration weight, h2 is the state weight, r1 is the first preset state index, r2 is the second preset state index, j0 is the number of monitoring cycles in which the engine vibration state is abnormal within the management cycle, and J is the number of monitoring cycles within the management cycle.

6. The engine fault detection and diagnosis method according to claim 5, characterized in that: When performing fluctuation analysis on the engine speed, the fluctuation coefficient is set to ZB, and the fluctuation coefficient ZB is compared and analyzed with the preset fluctuation threshold ZB1 to determine the engine speed fluctuation state. When the engine speed fluctuation state is abnormal, the engine fault level classification process is optimized, and the optimized first preset state index is set to r1', so that the first preset state index increases with the abnormal degree of the engine speed fluctuation state.

7. The engine fault detection and diagnosis method according to claim 6, characterized in that: The process of integrating the data of the transmission shaft stiffness and lubricating oil pressure is as follows: Compare and analyze the transmission shaft stiffness g0 with the preset stiffness g1 to construct a stiffness index, and compare and analyze the lubricating oil pressure y0 with the preset pressure y1 to construct a pressure index; The speed influence factor SY is constructed according to the stiffness index and the pressure index, and the analysis process of the engine speed fluctuation state is optimized according to the speed influence factor. The optimized preset fluctuation threshold is set to ZB2, so that the preset fluctuation threshold decreases with the increase of the speed influence factor.

8. The engine fault detection and diagnosis method according to claim 7, characterized in that: When extracting characteristic frequencies from the preprocessed engine vibration signal, the fundamental frequency is set to f1, and f1=R / 60 is set. Harmonic frequencies f2 and f3 are identified based on the fundamental frequency f1, and the amplitude of f1 is recorded as Af1, the amplitude of f2 is recorded as Af2, and the amplitude of f3 is recorded as Af3.

9. The engine fault detection and diagnosis method according to claim 8, characterized in that: The process of judging the engine vibration state based on the characteristic frequency is as follows: If Af2 / Af1>α or Af3 / Af1>α, the engine vibration state in the current monitoring period is determined to be abnormal; otherwise, the engine vibration state in the current monitoring period is determined to be normal, and α is a preset abnormal proportional coefficient.

10. The engine fault detection and diagnosis method according to claim 9, characterized in that: The flow fluctuation rate Qb of the coolant flow is calculated, and the flow fluctuation rate Qb and the average coolant flow rate Qp are jointly analyzed to determine the engine cooling state, where: If Qp<u1 or Qb>u2, the engine cooling state of the current management cycle is determined to be an abnormal state, otherwise, the engine cooling state of the current management cycle is determined to be a normal state; Among them, u1 is the first preset flow rate, and u2 is the second preset flow rate.

Images