A wind turbine tower top front and back displacement state estimation method and system

Abstract

The application provides a wind turbine tower top front and rear displacement state estimation method and system, which comprises the following steps: obtaining operation data of a wind turbine to be estimated, establishing a wind turbine state space model according to the operation data, and converting the wind turbine state space model into a discrete state space model; constructing a standard Kalman filter recursion framework according to the discrete state space model, predicting a prior state estimation value to obtain a prior state error covariance matrix; changing the prior state error covariance matrix into a square root covariance recursion mode to obtain a state error covariance matrix; calculating a measurement residual, improving a fixed measurement noise covariance matrix mode into an online correction mode to update the measurement noise covariance matrix online; calculating a filter gain, correcting the prior state estimation value by using the filter gain and the measurement residual; and updating the posterior state error covariance matrix to obtain a tower top front and rear displacement state estimation value. By this mode, the tower top front and rear displacement estimation accuracy can be improved, and the numerical stability can be enhanced.

Description

Technical Field

[0001] This application relates to the field of wind turbine technology, and in particular to a method and system for estimating the front and rear displacement state of the wind turbine tower top. Background Technology

[0002] As the capacity of individual wind turbine units continues to increase, the rotor diameter and tower height also increase, and the tower flexibility becomes more pronounced. Under the influence of factors such as turbulent wind, wind shear, pitch control, and transmission chain coupling, wind turbine units are prone to significant forward and backward vibrations. These vibrations not only affect the structural safety of the nacelle and tower but also lead to increased fatigue loads, ultimately impacting the long-term operational reliability of the wind turbine unit.

[0003] The forward and backward displacement of the tower top is an important state variable characterizing the forward and backward vibration state of the wind turbine tower. In existing technologies, the forward and backward motion state of the tower top is usually obtained directly or indirectly by devices such as accelerometers and displacement sensors installed on the nacelle or tower. However, in actual operating environments, the installation and maintenance costs of sensors are high, the measurement signals are easily affected by noise, drift and equipment failure, and some key state variables are difficult to measure directly.

[0004] The Kalman filter algorithm can recursively estimate state variables that cannot be directly measured using system models and measurement data, making it suitable for wind turbine state estimation problems. However, the standard Kalman filter algorithm is highly dependent on the accuracy of the statistical characteristics of process noise and measurement noise. When the noise statistics are unknown or change with operating conditions, the estimation accuracy is prone to decrease. At the same time, during long-term recursive calculations, the error covariance matrix may become non-positive definite or exhibit numerical divergence, affecting the stability of the state estimation.

[0005] Therefore, there is an urgent need for a state estimation method for the front and rear displacements of wind turbine towers, which can improve the estimation accuracy of the front and rear displacements of the tower tops and enhance the numerical stability of the filtering recursion process under the conditions of uncertainty in the wind turbine model and changes in measurement noise. Summary of the Invention

[0006] This application provides a method and system for estimating the front and rear displacement states of the wind turbine tower top, so as to improve the estimation accuracy of the front and rear displacement of the tower top under the conditions of uncertainty of the wind turbine model and variation of measurement noise, and enhance the numerical stability of the filtering recursion process.

[0007] Firstly, this application provides a method for estimating the forward and backward displacement state of the wind turbine tower top, which is executed by a computing device. The computing device can be understood as a computer or similar device, and is not limited thereto in this application. The method includes:

[0008] The process involves acquiring operational data of the wind turbine to be estimated, including wind speed, generator torque, pitch angle, and generator speed. A state-space model of the wind turbine, incorporating the first-order forward and backward vibration modes of the tower, is established based on this data, and then converted into a discrete state-space model. A standard Kalman filter recursive framework is constructed based on the discrete state-space model. This framework is then used to predict prior state estimates and obtain the prior state error covariance matrix in the standard Kalman filter. Finally, the method for calculating the prior state error covariance matrix in the standard Kalman filter is changed from direct recursion to square root covariance. The state error covariance matrix is ​​obtained through a recursive method. The measurement residuals are calculated, and the method of fixing the measurement noise covariance matrix in the standard Kalman filter is improved to an online correction method based on the improved Sage-Hussa adaptive noise statistical estimation. The measurement noise covariance matrix is ​​updated online based on the measurement residuals and the state error covariance matrix. The filter gain is calculated based on the updated measurement noise covariance matrix, and the prior state estimate is corrected using the filter gain and the measurement residuals. The posterior state error covariance matrix is ​​updated, and the estimated values ​​of the tower top displacement state are obtained based on the updated posterior state error covariance matrix.

[0009] Through the above methods, this application obtains the operating data and structural parameters of the wind turbine to be estimated, establishes a state-space model of the wind turbine including the first-order front and rear vibration modes of the tower, and discretizes the state-space model. Based on the discrete state-space model, an adaptive square-root Kalman filter is constructed. The numerical stability of the state error covariance matrix is ​​maintained by recursion of the square-root covariance. Combined with the improved Sage-Husa adaptive filtering method, the measurement noise covariance matrix is ​​corrected online, and the front and rear displacement state of the wind turbine tower is recursively estimated to obtain the estimated value of the front and rear displacement of the tower top. This application can improve the estimation accuracy and filtering stability of the front and rear displacement state of the tower top while reducing the dependence on the front and rear motion sensors of the tower top, providing reliable state variables for wind turbine tower vibration state monitoring, operation safety assessment, and subsequent controller design.

[0010] In the aforementioned method for estimating the forward and backward displacement state of the wind turbine tower, a state-space model of the wind turbine, including the first-order forward and backward vibration modes of the tower, is established based on operating data. This state-space model is then converted into a discrete state-space model, including:

[0011] ;

[0012] in, This represents the state vector at the k-th sampling time. This represents the measurement vector at the k-th sampling time. This represents the input vector at the (k-1)th sampling time. Represents the discrete state transition matrix. This represents the discrete input matrix at time k-1. This represents the measurement matrix at time k-1. Indicates process noise. Represents measurement noise; input vector u Including wind speed, generator torque, and blade pitch angle; measurement vectors This includes generator speed.

[0013] In the aforementioned method for estimating the displacement state of the wind turbine tower top, the prior state estimate is predicted based on the standard Kalman filter recursive framework, and the prior state error covariance matrix in the standard Kalman filter is obtained, including:

[0014] Based on the standard Kalman filter recursive framework, the posterior state estimate at the (k-1)th sampling time is used... Predict the prior state estimate at the k-th sampling time. And obtain the prior state error covariance matrix in standard Kalman filtering. ,Right now:

[0015] ;

[0016] in, This represents the prior state estimate at the k-th sampling time. This represents the posterior state estimate at the (k-1)th sampling time. Let the prior state error covariance matrix be represented at the k-th sampling time. Let represent the posterior state error covariance matrix at the (k-1)th sampling time. This represents the process noise covariance matrix.

[0017] In the aforementioned method for estimating the displacement state of the wind turbine tower top, the prior state error covariance matrix in the standard Kalman filter is changed from a direct recursive method to a square root covariance recursive method to obtain the state error covariance matrix, which includes:

[0018] The state error covariance matrix is ​​directly recursively derived in the standard Kalman filter. and The previous method was improved to a square root covariance recursive method, which expresses the state error covariance matrix as:

[0019] ;

[0020] in, This represents the state error covariance matrix at the k-th sampling time. This represents the square root matrix of the state error covariance at the k-th sampling time. express The transpose of the matrix; updating the square root of the state error covariance matrix during the filtering recursion process. This replaces the original directly recursive state error covariance matrix of the Kalman filter. .

[0021] Using the above method, the square root covariance recursion method of this application is used to replace the direct recursion of the prior state error covariance matrix in standard Kalman filtering. and the posterior state error covariance matrix This is done in a way that maintains the symmetry and positive definiteness of the state error covariance matrix.

[0022] In the aforementioned method for estimating the displacement state of the wind turbine tower top, the measurement residuals are calculated. The method of fixing the measurement noise covariance matrix in standard Kalman filtering is improved to an online correction method based on improved Sage-Hussa adaptive noise statistical estimation. The measurement noise covariance matrix is ​​updated online based on the measurement residuals and the state error covariance matrix, including:

[0023] Based on the measurement vector at the kth sampling time and predicted output value Calculate the measurement residuals:

[0024] ;

[0025] in, This represents the measurement residual at the k-th sampling time. This represents the measurement vector at the k-th sampling time. Represents the measurement matrix. Represents the prior state estimate at the k-th sampling time; the fixed measurement noise covariance matrix R in the standard Kalman filter... k The original method was improved to an online correction method based on improved Sage-Hussa adaptive noise statistical estimation, which updates the measurement noise covariance matrix online according to the measurement residuals.

[0026] ;

[0027] ;

[0028] in, This represents the measurement noise covariance matrix updated at the k-th sampling time. This represents the measurement noise covariance matrix at the k-th sampling time. Let represent the adaptive weighting coefficient, b represent the forgetting factor, and satisfy 0. <b<1, Indicates measurement residuals, This represents the transpose of the measurement residual. Represents the measurement matrix. Let the prior state error covariance matrix be represented at the k-th sampling time. This indicates the transpose of the measurement matrix;

[0029] Improved Sage-Husa Adaptive Noise Statistical Estimation Method Based on Measurement Residues For the measurement noise covariance matrix Online corrections are performed to adapt the filter to changes in noise statistical characteristics during wind turbine operation.

[0030] In the aforementioned method for estimating the displacement state of the wind turbine tower top, the filter gain is calculated based on the updated measurement noise covariance matrix, and the prior state estimate is corrected using the filter gain and measurement residuals, including:

[0031] Based on the updated measurement noise covariance matrix Calculate the filter gain:

[0032] ;

[0033] And utilize the filter gain and measurement residuals Prior state estimates Perform correction:

[0034] ;

[0035] Update the posterior state error covariance matrix:

[0036] ;

[0037] in, This represents the filter gain at the k-th sampling time. Let P represent the posterior state estimate at the k-th sampling time. k ⁺ represents the posterior state error covariance matrix at the k-th sampling time, and I represents the identity matrix.

[0038] The aforementioned method for estimating the displacement state of the wind turbine tower top also includes adaptive correction of the process noise covariance matrix. The update formula for the process noise covariance matrix is ​​as follows:

[0039] ;

[0040] in, This represents the process noise covariance matrix updated at the k-th sampling time. This represents the process noise covariance matrix at the (k-1)th sampling time. Indicates the adaptive weighting coefficients. Indicates the filter gain. Indicates measurement residuals, Let represent the posterior state error covariance matrix at the k-th sampling time. Let represent the posterior state error covariance matrix at the (k-1)th sampling time.

[0041] In a second aspect, this application provides a system for estimating the front and rear displacement states of a wind turbine tower top, used to execute the method of the first aspect of this application. The system includes: an acquisition module, a construction module, a recursive module, an improvement module, a calculation module, a correction module, and an update module.

[0042] The system comprises the following modules: an acquisition module for acquiring operational data of the wind turbine to be estimated, including wind speed, generator torque, pitch angle, and generator speed; a construction module for establishing a state-space model of the wind turbine including the first-order forward and backward vibration modes of the tower, and converting the state-space model into a discrete state-space model; a recursive module for constructing a standard Kalman filter recursive framework based on the discrete state-space model, predicting prior state estimates based on the standard Kalman filter recursive framework, and obtaining the prior state error covariance matrix in the standard Kalman filter; and an improvement module for changing the prior state error covariance matrix in the standard Kalman filter from a direct recursive method to a flattened method. The system employs a square root covariance recursive method to obtain the state error covariance matrix. A calculation module calculates the measurement residuals, improving upon the standard Kalman filtering method of fixing the measurement noise covariance matrix to an online correction method based on improved Sage-Hussa adaptive noise statistical estimation. This module updates the measurement noise covariance matrix online based on the measurement residuals and the state error covariance matrix. A correction module calculates the filter gain based on the updated measurement noise covariance matrix and uses the filter gain and measurement residuals to correct the prior state estimate. An update module updates the posterior state error covariance matrix, obtaining the estimated displacement state values ​​before and after the tower top based on the updated posterior state error covariance matrix.

[0043] Thirdly, this application also provides a computing device, comprising: a memory for storing program instructions; and a processor for calling the program instructions stored in the memory and executing the method described in the first aspect according to the obtained program instructions.

[0044] Fourthly, this application also provides a computer-readable storage medium storing computer-readable instructions, which, when read and executed by a computer, implement the method of the first aspect described above.

[0045] Fifthly, this application provides a computer program product including a computer program executable by a computer device, which, when run on the computer device, causes the computer device to perform the method described in the first aspect.

[0046] Beneficial Effects: Through the above methods, this application obtains the operating data and structural parameters of the wind turbine to be estimated, establishes a state-space model of the wind turbine including the first-order front and rear vibration modes of the tower, and discretizes the state-space model. Based on the discrete state-space model, an adaptive square-root Kalman filter is constructed. The numerical stability of the state error covariance matrix is ​​maintained by recursion of the square-root covariance. Combined with the improved Sage-Husa adaptive filtering method, the measurement noise covariance matrix is ​​corrected online, and the front and rear displacement state of the wind turbine tower is recursively estimated to obtain the estimated value of the front and rear displacement of the tower top. This application can improve the estimation accuracy and filtering stability of the front and rear displacement state of the tower top while reducing the dependence on the front and rear motion sensors of the tower top, providing reliable state variables for wind turbine tower vibration state monitoring, operation safety assessment, and subsequent controller design. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart illustrating a method for estimating the front and rear displacement states of a wind turbine tower top, provided in an embodiment of this application;

[0049] Figure 2 A simulation comparison diagram of the estimated and actual values ​​of the front and rear displacement of the wind turbine tower top using a Kalman filter in an embodiment of this application.

[0050] Figure 3 A simulation comparison of the estimated and actual values ​​of the front and rear displacement of the tower top under 14m / s GP_LLJ type turbulent wind, provided in an embodiment of this application, for a wind turbine tower top front and rear displacement state estimation system;

[0051] Figure 4 A simulation comparison of the estimated and actual values ​​of the front and rear displacement of a wind turbine tower top under 14 m / s SMOOTH type turbulent wind, provided in an embodiment of this application;

[0052] Figure 5 A simulation comparison of the estimated and actual values ​​of the front and rear displacement of a wind turbine tower top under 14 m / s NWTCUP type turbulent wind, provided in an embodiment of this application;

[0053] Figure 6 A schematic diagram of a wind turbine tower top displacement state estimation system provided in this application embodiment;

[0054] Figure 7 is a schematic structural diagram of a computing device provided in an embodiment of the present application. Detailed Description of the Embodiments

[0055] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0056] In the following embodiments of the present application, "and / or" describes the association relationship of associated objects, indicating that there may be three relationships. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B may be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one of the following" or similar expressions thereof refers to any combination of these items, including any combination of a single item (one) or a plurality of items (ones). For example, at least one (one) of a, b or c may represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, c may be a single one or a plurality of ones. The singular expressions "a", "an", "the", "the above", "said" and "this" are intended to also include expressions such as "one or more", unless the context clearly indicates otherwise. And, unless otherwise stated, ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish a plurality of objects, and are not used to limit the order, time sequence, priority or importance of the plurality of objects.

[0057] References in the specification of the present application to "one embodiment" or "some embodiments" and the like mean that a particular feature, structure or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, the appearances of the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in further some embodiments" and the like in various places in this specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in another manner. The terms "comprise", "include", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in another manner.

[0058] Embodiment 1

[0059] Embodiment 1 of the present application provides a method for estimating the front-rear displacement state of the top of a wind turbine, the method is executed by a computing device, and the computing device may be understood as a device such as a computer, which is not limited herein in the present application. The process of the method is as follows Figure 1 shown, comprising:

[0060] Step 101: Obtain the operating data of the wind turbine to be estimated. The operating data includes wind speed, generator torque, pitch angle and generator speed.

[0061] Step 102: Establish a state-space model of the wind turbine generator that includes the first-order front and rear vibration modes of the tower based on the operating data, and convert the state-space model into a discrete state-space model.

[0062] Specifically, converting the state-space model into a discrete state-space model includes:

[0063] ;

[0064] in, This represents the state vector at the k-th sampling time. This represents the measurement vector at the k-th sampling time. This represents the input vector at the (k-1)th sampling time. Represents the discrete state transition matrix. This represents the discrete input matrix at time k-1. This represents the measurement matrix at time k-1. Indicates process noise. Indicates measurement noise;

[0065] Input vector u Including wind speed, generator torque, and blade pitch angle; measurement vectors This includes generator speed.

[0066] In one possible implementation, the wind turbine state-space model is obtained by using FAST software with preset operating points and then further customized by setting additional operating points.

[0067] Step 103: Construct a standard Kalman filter recursive framework based on the discrete state space model, predict the prior state estimate based on the standard Kalman filter recursive framework, and obtain the prior state error covariance matrix in the standard Kalman filter.

[0068] Specifically, based on the standard Kalman filter recursive framework, the posterior state estimate at the (k-1)th sampling time is used... Predict the prior state estimate at the k-th sampling time. And obtain the prior state error covariance matrix in standard Kalman filtering. ,Right now:

[0069] ;

[0070] in, This represents the prior state estimate at the k-th sampling time. This represents the posterior state estimate at the (k-1)th sampling time. Let the prior state error covariance matrix be represented at the k-th sampling time. Let represent the posterior state error covariance matrix at the (k-1)th sampling time. This represents the process noise covariance matrix.

[0071] Step 104: Change the prior state error covariance matrix in the standard Kalman filter from the direct recursive method to the square root covariance recursive method to obtain the state error covariance matrix.

[0072] Specifically, the state error covariance matrix in standard Kalman filtering is directly recursively derived. and The previous method was improved to a square root covariance recursive method, which expresses the state error covariance matrix as:

[0073] ;

[0074] in, This represents the state error covariance matrix at the k-th sampling time. This represents the square root matrix of the state error covariance at the k-th sampling time. express The transpose of the matrix; updating the square root of the state error covariance matrix during the filtering recursion process. This replaces the original directly recursive state error covariance matrix of the Kalman filter. .

[0075] Step 105: Calculate the measurement residuals. Improve the fixed measurement noise covariance matrix method in standard Kalman filtering to an online correction method based on improved Sage-Husa adaptive noise statistical estimation. Update the measurement noise covariance matrix online according to the measurement residuals and state error covariance matrix.

[0076] Based on the measurement vector at the kth sampling time and predicted output value Calculate the measurement residuals:

[0077] ;

[0078] in, This represents the measurement residual at the k-th sampling time. This represents the measurement vector at the k-th sampling time. Represents the measurement matrix. This represents the prior state estimate at the k-th sampling time.

[0079] In standard Kalman filtering, the fixed measurement noise covariance matrix R is used. kThe original method was improved to an online correction method based on improved Sage-Hussa adaptive noise statistical estimation, which updates the measurement noise covariance matrix online according to the measurement residuals.

[0080] ;

[0081] ;

[0082] in, This represents the measurement noise covariance matrix updated at the k-th sampling time. This represents the measurement noise covariance matrix at the k-th sampling time. Let represent the adaptive weighting coefficient, b represent the forgetting factor, and satisfy 0. <b<1, Indicates measurement residuals, This represents the transpose of the measurement residual. Represents the measurement matrix. Let the prior state error covariance matrix be represented at the k-th sampling time. This indicates the transpose of the measurement matrix;

[0083] Improved Sage-Husa Adaptive Noise Statistical Estimation Method Based on Measurement Residues For the measurement noise covariance matrix Online corrections are performed to adapt the filter to changes in noise statistical characteristics during wind turbine operation.

[0084] Step 106: Calculate the filter gain based on the updated measurement noise covariance matrix, and use the filter gain and measurement residuals to correct the prior state estimate.

[0085] Specifically, based on the updated measurement noise covariance matrix Calculate the filter gain:

[0086] ;

[0087] And utilize the filter gain and measurement residuals Prior state estimates Perform correction:

[0088] ;

[0089] Update the posterior state error covariance matrix:

[0090] ;

[0091] in, This represents the filter gain at the k-th sampling time. Let P represent the posterior state estimate at the k-th sampling time.k ⁺ represents the posterior state error covariance matrix at the k-th sampling time, and I represents the identity matrix.

[0092] Step 107: Update the posterior state error covariance matrix and obtain the estimated values ​​of the tower top displacement state based on the updated posterior state error covariance matrix.

[0093] Specifically, from the corrected posterior state estimate The first-order forward and backward displacement state components of the tower are extracted and output as the estimated forward and backward displacement of the tower top, which is then used as the input of the subsequent wind turbine controller.

[0094] In one possible implementation, the estimated displacement values ​​of the tower top are output in time series form and used for monitoring the vibration status of the wind turbine tower, assessing the safety of tower operation, or as input signals to the wind turbine controller.

[0095] Example 2

[0096] This application provides a specific implementation method for estimating the front and rear displacement states of wind turbine towers based on the adaptive square root Kalman filter algorithm, according to Embodiment 1. The following specific implementation case illustrates the above-mentioned method for estimating the front and rear displacement states of wind turbine towers based on the adaptive square root Kalman filter algorithm.

[0097] The simulation object is a 5MW variable speed and variable pitch onshore wind turbine generator in the FAST software. Its detailed parameters are shown in Table 1.

[0098] Table 1 Technical Specifications of Wind Turbine Units

[0099] Wheel hub height (m) 90 Cut-in, rated, and cut-out wind velocities (m / s) 3，25，12 Impeller rated speed (rpm) 12.1 Transmission ratio 97 Rated power (MW) 5 Pitch angle range (°) 0~90 Maximum absolute pitch rate (° / s) 8 control Gear shifting, pitch shifting Cabin mass (kg) 240000 Tower mass (kg) 347460 Optimal tip speed ratio 7.55 Optimal wind energy utilization coefficient 0.48

[0100] S1. Obtain the operating data of the wind turbine to be estimated, including wind speed, generator torque, pitch angle and generator speed.

[0101] S2. Establish a state-space model of the wind turbine that includes the first-order front and rear vibration modes of the tower, and convert the state-space model into a discrete state-space model.

[0102] ;

[0103] in, This represents the state vector at the k-th sampling time. This represents the measurement vector at the k-th sampling time. This represents the input vector at the (k-1)th sampling time. Represents the discrete state transition matrix. This represents the discrete input matrix at time k-1. This represents the measurement matrix at time k-1. Indicates process noise. This indicates measurement noise.

[0104] S3. Based on the standard Kalman filter recursive framework, based on the posterior state estimate at the (k-1)th sampling time... Predict the prior state estimate at the k-th sampling time. And obtain the prior state error covariance matrix in standard Kalman filtering. ,Right now:

[0105] ;

[0106] in, This represents the prior state estimate at the k-th sampling time. This represents the posterior state estimate at the (k-1)th sampling time. Let the prior state error covariance matrix be represented at the k-th sampling time. Let represent the posterior state error covariance matrix at the (k-1)th sampling time. Represents the process noise covariance matrix;

[0107] S4. Directly recursively derive the state error covariance matrix in the standard Kalman filter. and The previous method was improved to a square root covariance recursive method, which expresses the state error covariance matrix as:

[0108] ;

[0109] in, This represents the state error covariance matrix at the k-th sampling time. This represents the square root matrix of the state error covariance at the k-th sampling time. express The transpose of the matrix; updating the square root of the state error covariance matrix during the filtering recursion process. This replaces the original directly recursive state error covariance matrix of the Kalman filter. .

[0110] S5. Based on the measurement vector at the k-th sampling time. and predicted output value Calculate the measurement residuals:

[0111] ;

[0112] in, This represents the measurement residual at the k-th sampling time. This represents the measurement vector at the k-th sampling time. Represents the measurement matrix. This represents the prior state estimate at the k-th sampling time.

[0113] S6. The fixed measurement noise covariance matrix R in the standard Kalman filter... k The previous method was improved to an online correction method based on improved Sage-Husa adaptive noise statistical estimation, which updates the measurement noise covariance matrix online according to the measurement residuals:

[0114] ;

[0115] ;

[0116] in, This represents the measurement noise covariance matrix updated at the k-th sampling time. This represents the measurement noise covariance matrix at the k-th sampling time. Let represent the adaptive weighting coefficient, b represent the forgetting factor, and satisfy 0. <b<1, Indicates measurement residuals, This represents the transpose of the measurement residual. Represents the measurement matrix. Let the prior state error covariance matrix be represented at the k-th sampling time. This represents the transpose of the measurement matrix.

[0117] S7. Based on the updated measurement noise covariance matrix Calculate the filter gain:

[0118] ;

[0119] And utilize the filter gain and measurement residuals Prior state estimates Perform correction:

[0120] ;

[0121] Simultaneously update the posterior state error covariance matrix:

[0122] ;

[0123] in, This represents the filter gain at the k-th sampling time. Let P represent the posterior state estimate at the k-th sampling time. k ⁺ represents the posterior state error covariance matrix at the k-th sampling time, and I represents the identity matrix;

[0124] S8. From the corrected posterior state estimate The first-order forward and backward displacement state components of the tower are extracted and output as the estimated forward and backward displacement of the tower top, which is then used as the input of the subsequent wind turbine controller.

[0125] like Figure 2 The figure shows a simulation comparison of the estimated and actual displacement values ​​of the tower top under the Kalman filter. Next, wind speed spectra of three turbulent wind types (GP_LLJ, SMOOTH, and NWTCUP) with a wind speed of 14 m / s are used to verify the effectiveness of the adaptive square root Kalman filter. The simulation comparison of the estimated and actual displacement values ​​of the tower top under the GP_LLJ type turbulent wind is shown below. Figure 3 As shown in the figure, the simulation comparison between the estimated and actual displacement values ​​of the tower top under SMOOTH-type turbulent wind is as follows. Figure 4 As shown in the figure, the simulation comparison between the estimated and actual displacement values ​​of the tower top under NWTCUP type turbulent wind is as follows. Figure 5 As shown; among them, GP_LLJ is a wind model used to simulate low-level jets, mainly used to study atmospheric circulation and weather systems; SMOOTH is a wind model used to simulate wind fields on smooth surfaces, suitable for urban environments and complex terrains; NWTCUP is a wind model developed by the National Wind Energy Technology Center (NWTC) in the United States for assessing wind energy resources and designing wind turbine generators.

[0126] The average deviation between the estimated and actual values ​​of different wind power spectrum models at a wind speed of 14 m / s is summarized in the table below:

[0127] Table 2. Average deviation of LASRKF algorithm under different wind power spectrum models

[0128] Average deviation (%) 10 5.3 5.4 5.4

[0129] like Figure 2 It can be seen that the estimation error of the standard Kalman filter algorithm is 10%, while the estimation error of the improved algorithm is about 5.3% on average, which shows that the improved algorithm estimates more accurately.

[0130] It can be seen that under the same wind speed, the errors are basically the same under different wind power spectrum models, indicating that different wind power spectrum models do not have a significant impact on the LA improvement algorithm. Because the GP_LLJ type turbulent wind estimation is more accurate, the GP_LLJ type turbulent wind will be used in the following discussion of the algorithm's effectiveness under different wind speeds.

[0131] The average deviation between the model's estimated values ​​and the actual values ​​obtained by comparing the LASRKF algorithm under different wind speeds is summarized in the table below:

[0132] Table 3 Average deviation of LASRKF algorithm at different wind speeds

[0133] Average deviation (%) 6.7 5.3 6.4 7

[0134] The table shows that the Least-squares Adaptive Square-Root Kalman Filter (LASRKF) algorithm has the smallest error and best performance in state estimation at a wind speed of 14 m / s. The average deviation within the range of rated wind speed to cutoff wind speed is about 5% to 7%, and the error gradually increases when approaching the rated wind speed and cutoff wind speed. It can be seen that the improved algorithm has higher accuracy than the Kalman filter and is more suitable for the standard cutoff wind speed of 14 m / s.

[0135] Example 3

[0136] Embodiment 3 of this application provides a system for estimating the front and rear displacement states of a wind turbine tower top, used to execute the methods provided in Embodiments 1 and 2 of this application. The system is as follows: Figure 6 As shown, it includes: an acquisition module, a construction module, a recursive module, an improvement module, a calculation module, a correction module, and an update module;

[0137] The acquisition module is used to acquire the operating data of the wind turbine to be estimated, including wind speed, generator torque, pitch angle and generator speed.

[0138] The module is used to build a state-space model of the wind turbine that includes the first-order front and rear vibration modes of the tower based on the operating data, and to convert the state-space model into a discrete state-space model.

[0139] The recursive module is used to construct a standard Kalman filter recursive framework based on the discrete state space model, predict the prior state estimate based on the standard Kalman filter recursive framework, and obtain the prior state error covariance matrix in the standard Kalman filter.

[0140] The improved module is used to change the prior state error covariance matrix in the standard Kalman filter from a direct recursive method to a square root covariance recursive method to obtain the state error covariance matrix.

[0141] The calculation module is used to calculate the measurement residuals. It improves the standard Kalman filtering method of fixing the measurement noise covariance matrix by using an online correction method based on the improved Sage-Hussa adaptive noise statistical estimation. The measurement noise covariance matrix is ​​updated online based on the measurement residuals and the state error covariance matrix.

[0142] The correction module is used to calculate the filter gain based on the updated measurement noise covariance matrix and to correct the prior state estimate using the filter gain and measurement residuals.

[0143] The update module is used to update the posterior state error covariance matrix and obtain the estimated values ​​of the tower top displacement state based on the updated posterior state error covariance matrix.

[0144] Example 4

[0145] After introducing a wind turbine tower top displacement state estimation system according to an exemplary embodiment of this application, the computing device of another exemplary embodiment of this application will be introduced next.

[0146] Those skilled in the art will understand that various aspects of this application can be implemented as a system, method, or program product. Therefore, various aspects of this application can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, collectively referred to herein as a "circuit," "module," or "system."

[0147] In some possible implementations, the computing device according to this application may include at least one processor and at least one memory. The memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the cross-device assisted interaction method for off-site clearing according to various exemplary embodiments of this application described above.

[0148] The following reference Figure 7 To describe a computing device 130 according to this embodiment of the present application. Figure 7 The computing device 130 shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this application. Figure 7 As shown, the computing device 130 is presented in the form of a general-purpose smart terminal (or Bluetooth headset). The components of the computing device 130 may include, but are not limited to: at least one processor 131, at least one memory 132, and a bus 133 connecting different system components (including memory 132 and processor 131).

[0149] Bus 133 represents one or more of several bus architectures, including a memory bus or memory controller, peripheral bus, processor, or local bus using any of the various bus architectures. Memory 132 may include readable media in the form of volatile memory, such as random access memory (RAM) 1321 and / or cache memory 1322, and may further include read-only memory (ROM) 1323. Memory 132 may also include a program / utility 1325 having a set (at least one) of program modules 1324, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0150] The computing device 130 can also communicate with one or more external devices 134 (e.g., keyboard, pointing device, etc.), and / or with any device that enables the computing device 130 to communicate with one or more other smart terminals (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 135. Furthermore, the computing device 130 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 136. As shown, network adapter 136 communicates with other modules used in the computing device 130 via bus 133. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with the computing device 130, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0151] In some possible implementations, various aspects of the wind turbine tower top displacement state estimation method provided in this application can also be implemented in the form of a program product, which includes a computer program. When the program product is run on a computer device, the computer program is used to cause the computer device to perform the steps in the wind turbine tower top displacement state estimation method according to various exemplary embodiments of this application as described above.

[0152] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0153] The program product for estimating the front and rear displacement states of the wind turbine tower top according to the embodiments of this application can be a portable compact disk read-only memory (CD-ROM) and include a computer program, and can run on a smart terminal. However, the program product of this application is not limited to this. In this document, the readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0154] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a readable computer program. Such propagated data signals may take various forms, including—but not limited to—electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0155] It should be noted that although several units or sub-units of the device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units.

[0156] Furthermore, although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0157] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable access frequency prediction device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable access frequency prediction device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0158] These computer program instructions may also be stored in a computer-readable storage medium capable of directing a computer or other programmable access predictive device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0159] These computer program instructions can also be loaded onto a computer or other programmable access device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable device for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0160] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0161] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for estimating the front and rear displacement state of the wind turbine tower top, characterized in that, include: Obtain the operating data of the wind turbine to be estimated, including wind speed, generator torque, pitch angle, and generator speed; A state-space model of the wind turbine generator, including the first-order front and rear vibration modes of the tower, is established based on the operating data, and the state-space model is converted into a discrete state-space model. Based on the discrete state-space model, a standard Kalman filter recursive framework is constructed. The prior state estimate is predicted based on the standard Kalman filter recursive framework, and the prior state error covariance matrix in the standard Kalman filter is obtained. The prior state error covariance matrix in the standard Kalman filter is obtained by changing the direct recursive method to the square root covariance recursive method. The measurement residuals are calculated, and the fixed measurement noise covariance matrix method in the standard Kalman filter is improved to an online correction method based on the improved Sage-Hussa adaptive noise statistical estimation. The measurement noise covariance matrix is ​​updated online according to the measurement residuals and the state error covariance matrix. The filter gain is calculated based on the updated measurement noise covariance matrix, and the prior state estimate is corrected using the filter gain and the measurement residual. Update the posterior state error covariance matrix, and obtain the estimated values ​​of the tower top displacement state based on the updated posterior state error covariance matrix.

2. The method according to claim 1, characterized in that, The step of establishing a wind turbine state-space model containing the first-order front and rear vibration modes of the tower based on the operating data, and converting the state-space model into a discrete state-space model, includes: ； in, This represents the state vector at the k-th sampling time. This represents the measurement vector at the k-th sampling time. This represents the input vector at the (k-1)th sampling time. Represents the discrete state transition matrix. This represents the discrete input matrix at time k-1. This represents the measurement matrix at time k-1. Indicates process noise. Indicates measurement noise; The input vector u Including wind speed, generator torque, and blade pitch angle; the measurement vector This includes generator speed.

3. The method according to claim 1, characterized in that, The step of predicting the prior state estimate based on the standard Kalman filter recursive framework and obtaining the prior state error covariance matrix in the standard Kalman filter includes: Based on the standard Kalman filter recursive framework, the posterior state estimate at the (k-1)th sampling time is used... Predict the prior state estimate at the k-th sampling time. And obtain the prior state error covariance matrix in standard Kalman filtering. ,Right now: ； in, This represents the prior state estimate at the k-th sampling time. This represents the posterior state estimate at the (k-1)th sampling time. Let the prior state error covariance matrix be represented at the k-th sampling time. Let represent the posterior state error covariance matrix at the (k-1)th sampling time. This represents the process noise covariance matrix.

4. The method according to claim 1, characterized in that, The step of changing the method of direct recursion for obtaining the prior state error covariance matrix in the standard Kalman filter to the method of square root covariance recursion, includes: The state error covariance matrix is ​​directly recursively derived in the standard Kalman filter. and The previous method was improved to a square root covariance recursive method, which expresses the state error covariance matrix as: ； in, This represents the state error covariance matrix at the k-th sampling time. This represents the square root matrix of the state error covariance at the k-th sampling time. express The transpose matrix; the square root matrix of the state error covariance is updated during the filtering recursion process. This replaces the original directly recursive state error covariance matrix of the Kalman filter. .

5. The method according to claim 1, characterized in that, The calculation of the measurement residual improves the standard Kalman filtering method of fixing the measurement noise covariance matrix to an online correction method based on improved Sage-Hussa adaptive noise statistical estimation. This involves updating the measurement noise covariance matrix online based on the measurement residual and the state error covariance matrix, including: Based on the measurement vector at the kth sampling time and predicted output value Calculate the measurement residuals: ； in, This represents the measurement residual at the k-th sampling time. This represents the measurement vector at the k-th sampling time. Represents the measurement matrix. This represents the prior state estimate at the k-th sampling time. In standard Kalman filtering, the fixed measurement noise covariance matrix R is used. k The method is improved to an online correction method based on improved Sage-Hussa adaptive noise statistical estimation, which updates the measurement noise covariance matrix online according to the measurement residuals; ； ； in, This represents the measurement noise covariance matrix updated at the k-th sampling time. This represents the measurement noise covariance matrix at the k-th sampling time. Let represent the adaptive weighting coefficient, b represent the forgetting factor, and satisfy 0. <b<1, Indicates measurement residuals, This represents the transpose of the measurement residual. Represents the measurement matrix. Let the prior state error covariance matrix be represented at the k-th sampling time. This indicates the transpose of the measurement matrix; The improved Sage-Husa adaptive noise statistical estimation method is based on the measurement residuals. For the measurement noise covariance matrix Online corrections are performed to adapt the filter to changes in noise statistical characteristics during wind turbine operation.

6. The method according to claim 1, characterized in that, The step of calculating the filter gain based on the updated measurement noise covariance matrix and correcting the prior state estimate using the filter gain and the measurement residual includes: Based on the updated measurement noise covariance matrix Calculate the filter gain: ； And utilize the filter gain and the measurement residual Prior state estimates Perform correction: ； Update the posterior state error covariance matrix: ； in, This represents the filter gain at the k-th sampling time. Let P represent the posterior state estimate at the k-th sampling time. k ⁺ represents the posterior state error covariance matrix at the k-th sampling time, and I represents the identity matrix.

7. The method according to claim 1, characterized in that, The method further includes adaptively correcting the process noise covariance matrix, wherein the update formula for the process noise covariance matrix is: ； in, This represents the process noise covariance matrix updated at the k-th sampling time. This represents the process noise covariance matrix at the (k-1)th sampling time. Indicates the adaptive weighting coefficients. Indicates the filter gain. Indicates measurement residuals, Let represent the posterior state error covariance matrix at the k-th sampling time. Let represent the posterior state error covariance matrix at the (k-1)th sampling time.

8. A system for estimating the front and rear displacement states of a wind turbine tower top, characterized in that, include: The acquisition module is used to acquire the operating data of the wind turbine to be estimated, including wind speed, generator torque, pitch angle and generator speed; The construction module is used to establish a wind turbine state space model containing the first-order front and rear vibration modes of the tower based on the operating data, and to convert the state space model into a discrete state space model. The recursive module is used to construct a standard Kalman filter recursive framework based on the discrete state space model, predict the prior state estimate based on the standard Kalman filter recursive framework, and obtain the prior state error covariance matrix in the standard Kalman filter. The improved module is used to change the prior state error covariance matrix in the standard Kalman filter from a direct recursive method to a square root covariance recursive method to obtain the state error covariance matrix. The calculation module is used to calculate the measurement residuals. It improves the standard Kalman filtering method by fixing the measurement noise covariance matrix, replacing it with an online correction method based on improved Sage-Hussa adaptive noise statistical estimation. The module updates the measurement noise covariance matrix online based on the measurement residuals and the state error covariance matrix. The correction module is used to calculate the filter gain based on the updated measurement noise covariance matrix, and to correct the prior state estimate using the filter gain and the measurement residual. The update module is used to update the posterior state error covariance matrix and obtain the estimated values ​​of the tower top displacement state based on the updated posterior state error covariance matrix.

9. A computing device, characterized in that, Its features include: Memory, used to store program instructions; A processor is configured to invoke program instructions stored in the memory and execute the method as described in any one of claims 1-7 according to the obtained program instructions.

10. A computer-readable storage medium, characterized in that, Includes computer-readable instructions that, when read and executed by a computer, cause the method as described in any one of claims 1 to 7 to be implemented.

Images