Method for monitoring and decoupling multiple parameters of rotating machinery blades based on FBG sensor

Abstract

The present application relates to the field of rotating machinery monitoring and optical fiber sensing technology, and provides a rotating machinery blade multi-parameter monitoring and decoupling method based on FBG sensors, comprising: measuring the temperature coefficient of the FBG sensor, and collecting the reflected spectrum center wavelength data of the FBG sensor when the temperature changes in real time; calculating the real-time temperature of the blade; decoupling the wavelength change data caused by the structural thermal strain of the strain FBG sensor at different temperatures according to the first wavelength relative transformation amount and the real-time temperature, obtaining the structural thermal strain data and the mechanical strain data of the strain FBG sensor; calculating the first load data and the load change amount of the blade through the blade state information, and decoupling the second load data. The present application realizes the multi-parameter monitoring and decoupling of temperature, thermal strain, mechanical strain and different direction loads through the cooperative work of the temperature FBG sensor and the strain FBG sensor, and only uses a sparse FBG sensor network.

Description

Technical Field

[0001] This invention relates to the fields of rotating machinery monitoring and fiber optic sensing technology, and in particular to a method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors. Background Technology

[0002] Major equipment, such as aero-engines and wind turbines, operate in harsh and variable environments, enduring constantly changing loads. This results in intense mechanical stresses on various subsystems and components, seriously threatening the structural health of these equipment. As one of the most heavily loaded parts of rotating mechanical blades in major equipment, the blade root's structural strength and stability are crucial to the safe operation of the equipment and require real-time monitoring.

[0003] Fiber Bragg grating sensors are affected by temperature changes during load monitoring. Furthermore, changes in external temperature can cause thermal expansion of the blade structure, leading to monitoring errors. The thermal expansion of the structure is generally much greater than that of the fiber optic material itself. During long-term monitoring, the range of ambient temperature variations is often large, significantly impacting the monitoring process. For example, the temperature of aircraft engine blades can reach hundreds or even thousands of degrees Celsius during service. Similarly, in cold regions, wind turbines sometimes heat the blades to prevent icing, resulting in significant temperature fluctuations.

[0004] Currently, load monitoring based on fiber Bragg grating sensors typically only considers the effect of temperature on the fiber itself, without considering the thermal strain of the structure caused by temperature, or only uses simple material structures to compensate for the thermal strain of the structure. This is difficult to replace the complex composite material structure of the blade. Moreover, the glass transition temperature of some composite materials is within the range of blade temperature variation, and their thermal strain curves are complex and difficult to compensate for with simple linear methods.

[0005] Due to factors such as changes in ambient temperature, the complexity of blade structure, various interferences during calibration, and the constantly changing structural state during service, existing technologies struggle to accurately decouple the loads on each blade. Therefore, a novel method for identifying and decoupling loads on rotating machinery blades in critical equipment is urgently needed to address these technical shortcomings. Summary of the Invention

[0006] To address the shortcomings of the existing technology, this invention provides a method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors, characterized by comprising:

[0007] S10. Measure the temperature coefficient of the FBG sensor and collect the center wavelength data of the reflection spectrum of the FBG sensor installed on the blade as the temperature changes in real time. The FBG sensor includes a temperature FBG sensor and a strain FBG sensor. The center wavelength data of the reflection spectrum includes the first wavelength relative change of the temperature FBG sensor and the second wavelength relative change of the strain FBG sensor.

[0008] S20. The real-time temperature of the blade is calculated by the relative change of the first wavelength of the temperature FBG sensor with temperature.

[0009] S30. Based on the first wavelength relative transformation and real-time temperature, decouple the wavelength change data caused by the structural thermal strain of the strain FBG sensor at different temperatures, and obtain the structural thermal strain data of the strain FBG sensor.

[0010] S40. Construct a structural thermal strain database using structural thermal strain data from strain FBG sensors at different temperatures, and fit the structural thermal strain formula of strain FBG sensors.

[0011] S50. Calibrate the strain FBG sensor, acquire the relative change of the second wavelength of the strain FBG sensor under different blade states under calibration conditions in real time, record the blade state information at the current moment, calculate the real-time temperature change and structural thermal strain formula based on the temperature FBG sensor, perform temperature compensation on the calibration data of the strain FBG sensor, and obtain the mechanical strain data of the strain FBG sensor.

[0012] S60. Calculate the first load data and load change of the blade using the blade status information;

[0013] S70. Combine the first load data and the load change with the mechanical strain data to construct a strain-load inverse algorithm and decouple the second load data.

[0014] In one embodiment, each blade includes at least four strain FBG sensors and four temperature FBG sensors. The central grid area of ​​the strain FBG sensors and the temperature FBG sensors must be perpendicular to the same horizontal plane as the center of the blade, and the strain FBG sensors are set at a preset angular interval.

[0015] In one embodiment, the relative change in the first wavelength of the FBG sensor can be expressed as:

[0016]

[0017] Among them, K Temp Let be the temperature coefficient of the FBG sensor, which is related to the thermal expansion coefficient and thermo-optic coefficient of the fiber grating, and ΔT be the change in ambient temperature. Then, the real-time temperature can be expressed as:

[0018]

[0019] Where T0 is the initial ambient temperature, λ Bt The first center wavelength data of the temperature FBG sensor, Δλ Bt The change in the first center wavelength of the temperature FBG sensor is given by k, where k represents different times. Temp_t Let be the temperature coefficient of the FBG sensor at time k.

[0020] In one embodiment, the relative change in the second wavelength of the strain FBG sensor during blade service can be expressed as:

[0021]

[0022] Where, λ Bs The second center wavelength data of the strain FBG sensor during blade service, Δλ Bs α represents the second wavelength change of the strain FBG sensor during blade service. fiber α is the thermal expansion coefficient of optical fiber. structure K is the coefficient of thermal expansion of the blade structure. Temp_s The temperature coefficient of the strain FBG sensor is given, where ΔT is the change in ambient temperature (K). ε Δε is the fiber strain coefficient. m This refers to the mechanical strain data experienced by the optical fiber.

[0023] In one embodiment, the relative change in the second wavelength of the strain FBG sensor caused by the structural thermal strain at different temperatures when the blade is at rest can be expressed as:

[0024]

[0025] The method for calculating the structural thermal strain data of the strain FBG sensor is as follows:

[0026]

[0027] In the formula, f(T) is the relative change of the second wavelength under the calibrated blade's stationary state. The structural thermal strain function, Δε, is fitted to the real-time temperature. t K represents the thermal strain data experienced by the optical fiber. ε The strain coefficient of the optical fiber.

[0028] In one embodiment, the formula for calculating the actual mechanical strain data of the FBG strain sensor during blade service is as follows:

[0029]

[0030] In one embodiment, the decoupling formula for the second load data in different directions is:

[0031]

[0032] Where G is the system transfer function. This is the first load data. Let be the load variations in n different directions, where n is less than i. The second wavelength relative change during the calibration of the i-th strain FBG sensor. Let be the relative change in the second wavelength of the i-th strain FBG sensor, [·] H Let be the conjugate transpose of matrix [·].

[0033] In one embodiment, S80, the second load data is corrected in real time based on an adaptive Kalman filter to obtain the corrected third load data in different directions.

[0034] In one embodiment, the system equation of the adaptive Kalman filter is:

[0035]

[0036] Among them, among them, This is a priori estimate of the third load data for the state at time k based on the system equations and the FBG sensor state at time k-1; u k-1 The system input value at time k-1 is specifically the difference between the relative changes in the second wavelength of the strain FBG sensor at times k-1 and k-2. A is the system matrix of the system equations, B is the input matrix of the system equations, and q... k-1 The system state noise at time k-1;

[0037] The prior error covariance matrix is:

[0038]

[0039] Among them, Q k-1 Let K be the state noise covariance matrix at time k-1, and Kalman gain be:

[0040]

[0041] Where C is the output matrix, R k-1 The noise covariance matrix is ​​measured at time k-1;

[0042] The posterior estimates of the state variables are:

[0043]

[0044] Among them, Y k For the output variable, i.e., the second load data; r k-1The covariance matrix of the posterior error of the measurement noise at time k-1 is:

[0045]

[0046] Where I is the identity matrix.

[0047] In one embodiment, the state noise, observation noise, and their covariance of the adaptive Kalman filter system at time k are expressed using the maximum a posteriori estimation principle as follows:

[0048]

[0049]

[0050]

[0051]

[0052]

[0053] Where m is the set window length.

[0054] Based on the above, compared with the prior art, the multi-parameter monitoring and decoupling method for rotating machinery blades based on FBG sensors provided by the present invention has the following beneficial effects:

[0055] 1. This invention achieves multi-parameter monitoring and decoupling of temperature, thermal strain, mechanical strain, and loads in different directions by using a temperature FBG sensor and a strain FBG sensor in synergy, with only a sparse FBG sensor network.

[0056] 2. This invention introduces an evaluation index for the relative change of center wavelength, which reduces the relative error between different optical fibers and has higher sensitivity and universality.

[0057] 3. This invention combines the least squares method with adaptive Kalman filtering, which can self-correct in the event of inaccurate structural parameters and changes in calibration state. It has high sensitivity, strong adaptability, and requires few parameters.

[0058] Other features and beneficial effects of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other beneficial effects of the invention can be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Unless otherwise specified, the positional relationships shown in the drawings in the following description are based on the direction in which the components are drawn in the figure.

[0060] Figure 1 A flowchart of a multi-parameter monitoring and decoupling method for rotating machinery blades based on FBG sensors provided for an embodiment of the present invention;

[0061] Figure 2 This is a side view of a wind turbine blade with an FBG sensor installed, provided in Embodiment 1 of the present invention.

[0062] Figure 3 This is a top view of a wind turbine blade with an FBG sensor installed, as provided in Embodiment 1 of the present invention.

[0063] Figure 4 This is a flowchart of a multi-parameter monitoring and decoupling method for rotating machinery blades based on FBG sensors, provided in Embodiment 2 of the present invention.

[0064] Figure label:

[0065] 1 blade, 11 blade roots, 2 temperature FBG sensors

[0066] 3-strain FBG sensor Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. The technical features designed in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0068] In the description of this invention, it should be noted that all terms used in this invention (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains, and should not be construed as limiting the invention; it should be further understood that the terms used in this invention should be understood to have the same meaning as those in the context of this specification and in the relevant field, and should not be understood in an idealized or overly formal sense, except as expressly defined in this invention.

[0069] A method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors, comprising:

[0070] S10. Measure the temperature coefficient of the FBG sensor and collect the center wavelength data of the reflection spectrum of the FBG sensor installed on the blade 1 as the temperature changes. The FBG sensor includes a temperature FBG sensor 2 and a strain FBG sensor 3. The center wavelength data of the reflection spectrum includes the first wavelength relative change of the temperature FBG sensor 2 and the second wavelength relative change of the strain FBG sensor 3.

[0071] S20. Calculate the real-time temperature of blade 1 by the relative change of the first wavelength of temperature FBG sensor 2 with temperature.

[0072] S30. Based on the first wavelength relative transformation and real-time temperature, decouple the wavelength change data caused by the structural thermal strain of strain FBG sensor 3 at different temperatures, and obtain the structural thermal strain data of strain FBG sensor 3.

[0073] S40. Construct a structural thermal strain database using structural thermal strain data from strain FBG sensor 3 at different temperatures, and fit the structural thermal strain formula of strain FBG sensor 3.

[0074] S50. Calibrate the strain FBG sensor 3, acquire the relative change of the second wavelength of the strain FBG sensor 3 under different states of blade 1 under calibration conditions in real time, and record the blade state information at the current moment. Based on the temperature FBG sensor 2, calculate the real-time temperature change and the structural thermal strain formula to perform temperature compensation on the calibration data of the strain FBG sensor 3, and obtain the mechanical strain data of the strain FBG sensor 3.

[0075] S60. Calculate the first load data and load change of blade 1 using the blade status information;

[0076] S70. Combine the first load data and the load change with the mechanical strain data to construct a strain-load inverse algorithm and decouple the second load data.

[0077] S10. Measure the temperature coefficient of the FBG sensor and collect the center wavelength data of the reflection spectrum of the FBG sensor installed on the blade 1 as the temperature changes. The FBG sensor includes a temperature FBG sensor 2 and a strain FBG sensor 3. The center wavelength data of the reflection spectrum includes the first relative change of the wavelength of the temperature FBG sensor 2 and the second relative change of the wavelength of the strain FBG sensor 3.

[0078] In practice, before installing the FBG sensor, the temperature coefficients of the strain FBG sensor 3 and the temperature FBG sensor 2 need to be measured. After the FBG sensor is installed on the blade 1 and put into service, the original center wavelengths of the strain FBG sensor 3 and the temperature FBG sensor 2 are recorded, and the relative change of the first wavelength of the temperature FBG sensor 2 and the relative change of the second wavelength of the strain FBG sensor 3 are obtained.

[0079] Preferably, such as Figure 2 As shown, each blade 1 includes at least four strain FBG sensors 3 and four temperature FBG sensors 2. The central grid area of ​​the strain FBG sensor 3 and the temperature FBG sensor 2 must be perpendicular to the same horizontal plane as the center of the blade 1. The strain FBG sensors 3 are set at a preset angular interval.

[0080] In practice, when monitoring and decoupling the load on wind turbine blades, the preset angle between strain FBG sensors is 90°, and each strain FBG sensor 3 is paired with one temperature FBG sensor 2. This is because if fewer than four strain FBG sensors 3 are installed on each wind turbine blade 1, the accuracy during load decoupling will be significantly reduced, making it impossible to obtain an accurate load reading. Therefore, each blade 1 must have at least four strain FBG sensors 3, located on the same cross-section of the blade 1.

[0081] It should be noted that the above installation method is suitable for wind turbine blades, but may not be suitable for rotating blades 1 of other machines. The placement of FBG sensors needs to be set according to the actual situation of the blades 1 of different machines, and is not limited to the above quantity and setting method. Furthermore, FBG sensors do not need to be located in a specific position or at a specific angle.

[0082] Before attaching the FBG sensor, the bonding surface needs to be sanded and the bonding area cleaned with anhydrous ethanol or acetone. Alternatively, for composite material blades 1 or blades 1 made of other materials, an embedded sensor method can be used to integrate the FBG sensor into the blade structure during manufacturing.

[0083] S20. Calculate the real-time temperature of blade 1 by the relative change of the first wavelength of temperature FBG sensor 2 with temperature.

[0084] The relative change in the first wavelength of the FBG sensor can be expressed as:

[0085]

[0086] Where, λ B Δλ is the center wavelength of the FBG sensor. B K represents the change in the center wavelength of the FBG sensor (here, the FBG sensor includes temperature FBG sensor 2 and strain FBG sensor 3). Temp Let be the temperature coefficient of the FBG sensor, which is related to the thermal expansion coefficient and thermo-optic coefficient of the fiber grating, and ΔT be the change in ambient temperature. Then, the real-time temperature can be expressed as:

[0087]

[0088] Where T0 is the initial ambient temperature, λ Bt The first center wavelength data of temperature FBG sensor 2, Δλ Bt The change in the first center wavelength of temperature FBG sensor 2 is given by t, where t represents different times, and K is the wavelength of the sensor. Temp_t Let t be the temperature coefficient of FBG sensor 2 at time t.

[0089] S30. Based on the first wavelength relative transformation and real-time temperature, decouple the wavelength change data caused by the structural thermal strain of strain FBG sensor 3 at different temperatures, and obtain the structural thermal strain data of strain FBG sensor 3.

[0090] S40. Construct a structural thermal strain database using structural thermal strain data from strain FBG sensor 3 at different temperatures, and fit the structural thermal strain formula of strain FBG sensor 3.

[0091] S50. Calibrate the strain FBG sensor 3, acquire the relative change of the second wavelength of the strain FBG sensor 3 under different states of blade 1 under calibration conditions in real time, and record the blade state information at the current moment. Based on the temperature FBG sensor 2, calculate the real-time temperature change and the structural thermal strain formula to perform temperature compensation on the calibration data of the strain FBG sensor 3, and obtain the mechanical strain data of the strain FBG sensor 3.

[0092] In one embodiment, the relative change in the second wavelength of the strain FBG sensor 3 during blade 1 service can be expressed as:

[0093]

[0094] Where, λ Bs The second center wavelength data of strain FBG sensor 3 during blade 1's service life is Δλ. Bs α represents the change in the second wavelength of the strain FBG sensor 3 during the service of blade 1. fiber α is the thermal expansion coefficient of optical fiber.Structure K is the coefficient of thermal expansion of the blade structure. Temp_s The temperature coefficient of the strain FBG sensor is given by ΔT, where ΔT is the change in ambient temperature (K). ε Δε is the fiber strain coefficient. m This refers to the mechanical strain data experienced by the optical fiber.

[0095] The relative change of the second wavelength of the strain FBG sensor 3 during service consists of three parts: real-time temperature, structural thermal strain data, and mechanical strain data. Since the real-time temperature has been obtained from the above steps, it is still necessary to calculate the structural thermal strain data and mechanical strain data.

[0096] Preferably, the structural thermal strain data is calculated when the wind turbine is stationary. At this time, the external wind speed is low and the wind speed and direction are relatively stable. The center wavelength data of the reflection spectrum of each FBG sensor in each blade 1 as a function of temperature change are collected by means of self-heating of the blade 1 or local heating of the bonded area. In order to ensure that the blade 1 structure is sufficiently heated and stable, the heating time for each measurement temperature is maintained for at least 5 minutes, and the calibration temperature range can include the ambient temperature inside the blade 1 during normal operation and the temperature of self-heating.

[0097] The relative change in the second wavelength of the FBG sensor 3 caused by the structural thermal strain at different temperatures when blade 1 is at rest can be expressed as:

[0098]

[0099] The method for calculating the structural thermal strain data of strain FBG sensor 3 is as follows:

[0100]

[0101] In the formula, f(T) is the relative change of the second wavelength under the stationary state of blade 1. The structural thermal strain function, Δε, is fitted to the real-time temperature. t K represents the thermal strain data experienced by the optical fiber. ε The strain coefficient of the optical fiber.

[0102] The above steps have calculated the real-time temperature and structural thermal strain data of the FBG sensor 3 under calibration conditions for blade 1. Given the relative change in the second wavelength, the formula for calculating the actual mechanical strain data of the FBG sensor 3 during blade 1's service life is as follows:

[0103]

[0104] After obtaining the mechanical strain data from strain FBG sensor 3, blade state information is collected, and load data in different directions of blade 1 are calculated, specifically:

[0105] S60. Calculate the first load data and load change of blade 1 using the blade 1 state information;

[0106] The recorded blade status information includes the center of gravity and weight of the rotating machinery blade 1, the cross-sectional position of the FBG sensor, and the current pitch angle, hub elevation angle, rotor azimuth angle, and blade tilt angle of blade 1. The theoretical loads in different directions are calculated by combining the blade status information with coordinate changes.

[0107] S70. Combine the first load data and the load change with the mechanical strain data to construct a strain-load inverse algorithm and decouple the second load data.

[0108] The decoupling formula for the second load data in different directions is:

[0109]

[0110] Where G is the system transfer function. For the second load data in n different directions, For the first load data in n different directions, Let be the load variations in n different directions, where n is less than i. The second wavelength relative change during the calibration of the i-th strain FBG sensor 3. Let [·] be the relative change in the second wavelength of the i-th strain FBG sensor 3. H Let be the conjugate transpose of matrix [·].

[0111] In practical implementation, the strain-load inverse calculation algorithm is as follows: For example, in a linear system, the theoretical swaying bending moment M of the rotating blade 1 of the wind turbine is obtained through the blade state information. x and theoretical waving bending moment M y Given input variable M x and M y and output variables Then, its corresponding system transfer function G can be determined, and we have:

[0112]

[0113] in, Let be the relative wavelength change caused by the mechanical strain measured by the i-th strain FBG sensor 3, where i is the number of strain FBGs on the same blade 1. The load change ΔM is obtained according to the least squares method and calibration conditions. x_c ΔM y_c and the relative wavelength change caused by mechanical strain under calibrated conditions. The system transfer function G is calculated as follows:

[0114]

[0115] In the formula, for The MP inverse, for measured data, is generally a full-rank row, and its calculation formula is:

[0116]

[0117] In the formula, for The conjugate transpose of .

[0118] The real-time swaying moment and flapping moment can be expressed as:

[0119]

[0120] This is the first load data.

[0121] It should be noted that the type of load on the rotating blade 1 needs to be determined specifically based on the specific mechanical blade 1, and is not limited to the theoretical oscillation bending moment M in this embodiment. x and theoretical waving bending moment M y It may also include other different types of loads, all of which can be decoupled using the method provided by this invention.

[0122] The specific working process is as follows: With blade 1 in a static state, the real-time temperature and structural thermal strain data of blade 1 are calculated. A structural thermal strain database is constructed using this data. After the structural thermal strain data is fitted, the real-time wavelength changes of the temperature FBG sensor 2 and strain FBG sensor 3 installed on blade 1 during service are measured. The real-time temperature change is calculated using temperature FBG sensor 2, and temperature compensation is applied to the calibration data to obtain the mechanical strain data from strain FBG sensor 3. Subsequently, the mechanical strain data, initial load data, and load change are combined to construct a strain-load inverse algorithm, decoupling loads in different directions.

[0123] Example 2

[0124] During the service of rotating machinery in major equipment, pitch, yaw, and vibration can all cause changes in the structural state, leading to certain errors in the loads obtained from state decoupling when the state is not completely consistent with the calibration state. To solve the problem of errors caused in the above situations, this invention also provides Embodiment Two, specifically a multi-parameter monitoring and decoupling method for rotating machinery blade 1 based on FBG sensors, which further includes:

[0125] S80. The second load data is corrected in real time based on the adaptive Kalman filter to obtain the corrected third load data in different directions.

[0126] The system equations for the adaptive Kalman filter are:

[0127]

[0128] in, This is a priori estimate of the third load data for the state at time k based on the system equations and the FBG sensor state at time k-1; u k-1 Let q be the system input value at time k-1, specifically the difference between the relative changes in the second wavelength of the strain FBG sensor at times k-1 and k-2. Let A be the system matrix of the system equations, B be the input matrix of the system equations, and q be the input matrix of the system equations. k-1 The system state noise at time k-1 is given.

[0129] In practice, A and B are the system matrix and input matrix of the system equations constructed based on the actual system, respectively. The actual system is specifically determined according to the different physical models of the FBG sensor installed on different blades 1. A and B are the equation parameters of the physical model. For example, in the wind turbine blade model, A can be the identity matrix I and B can be the system transfer function G. However, it should be noted that those skilled in the art can select A and B according to the actual model and are not limited to the values ​​limited above. For the corrected third load data, u k-1 The difference in the relative change of the second wavelength of strain FBG sensor 3 at times k-1 and k-2.

[0130] The prior error covariance matrix is:

[0131]

[0132] Among them, A T Let Q be the transpose of A. k-1 Let K be the state noise covariance matrix at time k-1, and Kalman gain be:

[0133]

[0134] Where C is the output matrix, C T Let R be the conjugate transpose of C. k-1 The noise covariance matrix is ​​measured at time k-1;

[0135] The posterior estimates of the state variables are:

[0136]

[0137] Among them, Y k For the output variable, i.e., the second load data; r k-1 Measure the noise at time k-1;

[0138] The posterior error covariance matrix is:

[0139]

[0140] Where I is the identity matrix, which is a matrix with 1s on the diagonal and 0s on the rest.

[0141] When a wind turbine is running, the ambient noise changes constantly and the noise parameters are difficult to observe directly. To address the system non-convergence problem caused by unknown or inaccurate state noise and observed noise, maximum a posteriori estimation is used to represent the noise parameters. Furthermore, by limiting the window length of the maximum a posteriori estimation, the influence of excessively distant data states on the parameters is reduced, thereby enhancing its dynamic characteristics.

[0142] The state noise, observation noise, and their covariance of the adaptive Kalman filter system at time k are expressed by the maximum a posteriori estimation principle as follows:

[0143]

[0144]

[0145]

[0146]

[0147]

[0148] Where, q k Let Q be the system state noise at time k. k Let R be the system state noise covariance matrix at time k, rk be the measurement noise at time k, and R be the system state noise covariance matrix at time k. k Let m be the noise covariance matrix measured at time k, m be the set window length, and j be used for counting within the loop. for The conjugate transpose matrix.

[0149] When k is less than the window length m, all data prior to k are averaged to calculate the noise parameters. When k is greater than m, to prevent outdated data from having an excessive impact on the current state, only data within the window length is used for calculation. Unlike existing technologies that use empirical values ​​to calculate and evaluate noise parameters, this invention uses maximum a posteriori estimation to calculate the noise parameters involved, employing statistical methods and replacing existing results with past data. This allows for real-time noise analysis based on system changes, resulting in more accurate noise parameters compared to empirical values.

[0150] Example 3

[0151] The multi-parameter monitoring and decoupling method for rotating machinery blades based on FBG sensors provided in Embodiments 1 and 2 above can be readily applied to the multi-parameter measurement and decoupling of loads on rotating machinery blades of various major equipment, including but not limited to wind turbine blades and aero-engine fan blades. It should be noted that when applying the multi-parameter monitoring and decoupling method for rotating machinery blades based on FBG sensors provided in Embodiments 1 and 2 to aero-engine fan blades and other major equipment rotating machinery blades, the placement and number of FBG sensors need to be set according to the actual situation, and are not limited to the placement and number specified in Embodiment 1. Simultaneously, the blade load type also needs to be adjusted according to the actual situation.

[0152] Furthermore, those skilled in the art should understand that although many problems exist in the prior art, each embodiment or technical solution of the present invention can be improved in only one or a few aspects, without necessarily solving all the technical problems listed in the prior art or the background art simultaneously. Those skilled in the art should understand that any content not mentioned in a claim should not be construed as a limitation on that claim.

[0153] Although this document frequently uses terms such as FBG sensor, temperature coefficient, center wavelength data of reflectance spectrum, temperature FBG sensor, strain FBG sensor, relative change of first wavelength, relative change of second wavelength, blade, real-time temperature, structural thermal strain data, blade condition information, mechanical strain data, first load data, load change, and second load data, the possibility of using other terms is not excluded. The use of these terms is merely for the convenience of describing and explaining the essence of the invention; interpreting them as any additional limitation would contradict the spirit of the invention. The terms "first," "second," etc. (if present) in the specification, claims, and accompanying drawings of the embodiments of the invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for FBG sensor based multi-parameter monitoring and decoupling of rotating machinery blades, characterized in that ,include: S10. Measure the temperature coefficient of the FBG sensor and collect the center wavelength data of the reflection spectrum of the FBG sensor installed on the blade as the temperature changes in real time. The FBG sensor includes a temperature FBG sensor and a strain FBG sensor. The center wavelength data of the reflection spectrum includes the first relative change of the wavelength of the temperature FBG sensor and the second relative change of the wavelength of the strain FBG sensor. S20. The real-time temperature of the blade is calculated by the relative change of the first wavelength of the temperature FBG sensor with temperature. S30. Based on the first relative wavelength change and the real-time temperature, decouple the wavelength change data caused by the structural thermal strain of the strain FBG sensor at different temperatures to obtain the structural thermal strain data of the strain FBG sensor. S40. Construct a structural thermal strain database using the structural thermal strain data of the strain FBG sensor at different temperatures, and fit the structural thermal strain formula of the strain FBG sensor. S50. The strain FBG sensor is calibrated, and the relative change of the second wavelength of the strain FBG sensor under different blade states under calibration conditions is acquired in real time. The blade state information at the current moment is recorded. The real-time temperature change is calculated based on the temperature FBG sensor and the structural thermal strain formula. Temperature compensation is performed on the calibration data of the strain FBG sensor to obtain the mechanical strain data of the strain FBG sensor. S60. Calculate the first load data and load change of the blade using the blade status information; S70. Combine the first load data and the load change with the mechanical strain data to construct a strain-load inverse calculation algorithm and decouple the second load data. The relative change in the first wavelength of the FBG sensor can be expressed as: wherein is the center wavelength of the FBG sensor, is the center wavelength variation of the FBG sensor, is the temperature coefficient of the FBG sensor, which is related to the thermal expansion coefficient and the thermo-optic coefficient of the fiber grating, is the ambient temperature variation, then the real-time temperature can be expressed as: wherein, is the initial temperature of the environment, is the first center wavelength data of the temperature FBG sensor, is the first center wavelength variation of the temperature FBG sensor, k represents different time points, is the temperature coefficient of the temperature FBG sensor at k time point; The relative change in the second wavelength of the strain FBG sensor during blade service can be expressed as: in, This refers to the second center wavelength data of the strain FBG sensor used during blade service. This refers to the second wavelength change of the strain FBG sensor during blade service. The coefficient of thermal expansion of optical fiber. The coefficient of thermal expansion of the blade structure. The temperature coefficient of the strain FBG sensor is... This refers to the change in ambient temperature. The fiber strain coefficient The mechanical strain data experienced by the optical fiber; The relative change in the second wavelength of the strain FBG sensor caused by the structural thermal strain at different temperatures when the blade is at rest can be expressed as: The structural thermal strain data of the strain FBG sensor is calculated as follows: In the formula, It is the relative change of the second wavelength when the blade is at rest. The structural thermal strain function is fitted to the real-time temperature. This is the thermal strain data experienced by the optical fiber. The strain coefficient of the optical fiber.

2. The method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors according to claim 1, characterized in that, Each blade includes at least four strain FBG sensors and four temperature FBG sensors. The central grid area of ​​the strain FBG sensor and the temperature FBG sensor must be perpendicular to the same horizontal plane as the center of the blade. The strain FBG sensors are arranged at a preset angular interval.

3. The method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors according to claim 1, characterized in that, The formula for calculating the actual mechanical strain data of the strain FBG sensor during blade service is as follows: 。 4. The method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors according to claim 1, characterized in that, The decoupling formulas for the second load data in different directions are: Where G is the system transfer function. For the first load data, Let n be the load variations in different directions; n is less than i, The second wavelength relative change during the calibration of the i-th strain FBG sensor. Let i be the relative change of the second wavelength of the i-th strain FBG sensor. For matrix The conjugate transpose of .

5. The method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors according to claim 1, characterized in that, Also includes: S80. The second load data is corrected in real time based on the adaptive Kalman filter to obtain the corrected third load data in different directions.

6. The method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors according to claim 5, characterized in that, The system equations for the adaptive Kalman filter are: in, Based on the system equations and The prior estimate of the third load data at time k based on the FBG sensor state at time k; for The system input value at a given time is specifically the strain FBG sensor at... and The difference in the relative change of the second wavelength at time t, where A is the system matrix of the system equations and B is the input matrix of the system equations. for Constant system state noise; The prior error covariance matrix is: in, Let A be the transpose of A. for The state noise covariance matrix at time step is given by the Kalman gain: Where C is the output matrix. Let C be the transpose of the matrix. for The noise covariance matrix is ​​measured at any given time. The posterior estimates of the state variables are: in, The output variable is specifically the second load data; for Measure noise at all times; The posterior error covariance matrix is: Where I is the identity matrix.

7. The method for multi-parameter monitoring and decoupling of rotating machinery blades based on FBG sensors according to claim 6, characterized in that, The state noise, observation noise, and their covariance of the adaptive Kalman filter system at time k are expressed by the maximum a posteriori estimation principle as follows: in, Let k be the system state noise at time k. Let k be the system state noise covariance matrix at time k. Noise is measured at time k. Let m be the noise covariance matrix measured at time k, m be the set window length, and j be used for counting within the loop.

Images