Response-driven power system broadband oscillation modal parameter decomposition method and device

Abstract

The invention discloses a response-driven power system broadband oscillation modal parameter decomposition method and device, and relates to the technical field of power. The method comprises the following steps: acquiring a signal of a power system related to broadband oscillation; performing noise reduction processing on the signal, related to the broadband oscillation, of the power system through an S-G filter to obtain a signal to be identified; establishing a difference equation of the signal to be identified; expressing the difference equation as a matrix form to obtain a matrix equation; obtaining a normal equation matrix of a TLS-Prony algorithm according to the matrix equation; solving the normal equation matrix to obtain a root of a characteristic polynomial; and obtaining modal parameters of the signal to be identified according to the root of the characteristic polynomial. According to the method, the influence of noise on parameter estimation can be effectively reduced, the precision of parameter estimation is improved, and abnormal values or noise components in data can be effectively identified and eliminated, so that the problem of matrix ill-conditioned caused by a traditional prony algorithm is avoided.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a method and device for decomposing wide-band oscillation modal parameters of a response-driven electric power system. Background Art

[0002] At present, the theoretical analysis methods for wind power grid-connected system oscillation parameter identification are mainly divided into two categories. One is based on mathematical models, and commonly used methods include eigenvalue analysis and Kalman filtering. These methods rely on accurate system models, but the "double high" characteristics of the power system are becoming more and more obvious, and the operation status of the power grid is complex and changeable. It is quite difficult to establish an accurate research model, and the limitations of analysis methods based on specific models are becoming more and more obvious. The second is based on data-driven methods, mainly including (improved) fast Fourier transform (FFT), Prony analysis, wavelet transform, Hilbert-Huang transform (HHT), etc. In order to improve the accuracy of oscillation identification, related studies often improve and integrate multiple algorithms, or pre-process the data to achieve noise reduction or avoid modal aliasing.

[0003] Some methods combine singular value decomposition, variational mode decomposition (VMD) and second-order blind identification to improve the Prony algorithm, and verify through measured data that the fusion algorithm has accurate identification and strong noise resistance, but these methods are not suitable for analyzing time-varying signals. HHT is an algorithm that combines empirical mode decomposition (EMD) with Hilbert transform (HT). In the early stage of a fault, when the oscillation is weak, it can accurately identify the oscillation frequency, but EMD has serious modal aliasing problems. In this regard, some methods use improved algorithms for modal decomposition and calculate the Teager energy operator to obtain the instantaneous parameters of the oscillation, but the parameter setting of the improved algorithm relies too much on experience. In addition, the continuous development of artificial intelligence and data mining capabilities has provided new ideas for oscillation research. However, existing studies are mostly based on training and learning based on simulation model data, and their applicability in actual power grids remains to be verified.

[0004] In summary, the "double high" characteristics of the current power system are becoming more and more obvious, the operating status of the power grid is complex and changeable, it is quite difficult to establish an accurate research model, and the limitations of the oscillation parameter identification method based on mathematical models are becoming more and more obvious; and the existing data-driven identification methods have problems such as excessive reliance on experience in parameter setting; in addition, existing research is mostly based on simulation model data for training and learning, and its applicability in actual power grids remains to be verified.

[0005] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No description herein is admitted to be prior art by inclusion in this section. Summary of the invention

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a method and device for decomposing modal parameters of a response-driven power system wide-band oscillation, which can solve at least part of the above technical problems.

[0007] An embodiment of the present invention provides a method for decomposing modal parameters of a response-driven power system wide-band oscillation, comprising:

[0008] Acquire signals related to broadband oscillations in power systems;

[0009] Performing noise reduction processing on the signal related to the broadband oscillation of the power system by using an SG filter to obtain a signal to be identified;

[0010] Establishing a differential equation of the signal to be identified;

[0011] Expressing the difference equation in matrix form to obtain a matrix equation;

[0012] Obtaining the normal equation matrix of the TLS-Prony algorithm according to the matrix equation;

[0013] Solving the normal equation matrix to obtain roots of the characteristic polynomial;

[0014] According to the roots of the characteristic polynomial, the modal parameters of the signal to be identified are obtained.

[0015] In some embodiments, the method further comprises:

[0016] The broadband oscillation state of the power system is determined according to the modal parameters of the signal to be identified.

[0017] In some embodiments, performing noise reduction processing on the signal through an SG filter to obtain a signal to be identified includes:

[0018] Filter the signal using a combination of different window lengths and polynomial degrees;

[0019] For each of the combinations, calculating the waveform similarity coefficient and mean square error of the signal before and after filtering;

[0020] Determine the parameters of the SG filter based on the waveform similarity coefficient and the mean square error;

[0021] The SG filter with the parameters set is used to perform noise reduction processing on the signal related to the broadband oscillation to obtain a signal to be identified.

[0022] In some embodiments, establishing the differential equation of the signal to be identified includes:

[0023] Establishing a fitting function of the signal to be identified;

[0024] According to the fitting function, a difference equation is established;

[0025] Establishing an error function between the signal to be identified and the approximate signal obtained by the differential equation;

[0026] With the goal of minimizing the error function, the system parameters of the difference equation are optimized by the TS algorithm to obtain the optimal system parameters;

[0027] The system parameters of the differential equation are set to the optimal system parameters to obtain the differential equation of the signal to be identified.

[0028] In some embodiments, the normal equation matrix of the TLS-Prony algorithm is obtained according to the matrix equation, including:

[0029] Applying singular value decomposition to the matrix equation to obtain a decomposition result;

[0030] According to the decomposition results, the normal equation matrix of the TLS-Prony algorithm is established.

[0031] In some embodiments, solving the normal equation matrix to obtain the roots of the characteristic polynomial includes:

[0032] According to the normal equation matrix, construct coefficients of characteristic polynomials;

[0033] The characteristic polynomial is solved according to the coefficients of the characteristic polynomial to obtain the roots of the characteristic polynomial.

[0034] An embodiment of the present invention provides a response-driven power system broadband oscillation modal parameter decomposition device, comprising:

[0035] An acquisition module, used for acquiring signals related to the wide-band oscillation of the power system;

[0036] A noise reduction module, used for performing noise reduction processing on the signal related to the broadband oscillation of the power system through an SG filter to obtain a signal to be identified;

[0037] An establishing module, used for establishing a differential equation of the signal to be identified;

[0038] A first conversion module, used for expressing the differential equation in a matrix form to obtain a matrix equation;

[0039] A second conversion module, used for obtaining a normal equation matrix of the TLS-Prony algorithm according to the matrix equation;

[0040] A first solving module, used for solving the normal equation matrix to obtain the roots of the characteristic polynomial;

[0041] The second solving module is used to obtain the modal parameters of the signal to be identified according to the roots of the characteristic polynomial.

[0042] In some embodiments, the apparatus further comprises:

[0043] The determination module is used to determine the broadband oscillation state of the power system according to the modal parameters of the signal to be identified.

[0044] In some embodiments, the noise reduction module is specifically used to:

[0045] Filter the signal using a combination of different window lengths and polynomial degrees;

[0046] For each of the combinations, calculating the waveform similarity coefficient and mean square error of the signal before and after filtering;

[0047] Determine the parameters of the SG filter based on the waveform similarity coefficient and the mean square error;

[0048] The SG filter with the parameters set is used to perform noise reduction processing on the signal related to the broadband oscillation to obtain a signal to be identified.

[0049] In some embodiments, the establishment module is specifically used to:

[0050] Establishing a fitting function of the signal to be identified;

[0051] According to the fitting function, a difference equation is established;

[0052] Establishing an error function between the signal to be identified and the approximate signal obtained by the differential equation;

[0053] With the goal of minimizing the error function, the system parameters of the difference equation are optimized by the TS algorithm to obtain the optimal system parameters;

[0054] The system parameters of the differential equation are set to the optimal system parameters to obtain the differential equation of the signal to be identified.

[0055] In some embodiments, the second conversion module is specifically configured to:

[0056] Applying singular value decomposition to the matrix equation to obtain a decomposition result;

[0057] According to the decomposition results, the normal equation matrix of the TLS-Prony algorithm is established.

[0058] In some embodiments, the first solution module is specifically used to:

[0059] According to the normal equation matrix, construct coefficients of characteristic polynomials;

[0060] The characteristic polynomial is solved according to the coefficients of the characteristic polynomial to obtain the roots of the characteristic polynomial.

[0061] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the response-driven power system wide-band oscillation modal parameter decomposition method described in any of the above embodiments is implemented.

[0062] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the response-driven power system wide-band oscillation modal parameter decomposition method described in any of the above embodiments.

[0063] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method for decomposing modal parameters of a response-driven power system wide-band oscillation is implemented as described in any of the above embodiments.

[0064] The response-driven power system broadband oscillation modal parameter decomposition method and device provided by the embodiment of the present invention firstly uses the Savizky-Golay filter to realize data smoothing; then establishes a differential equation to find the minimum error function of a given identification signal and an approximate signal; secondly, obtains the TS solution through the TS algorithm, constructs a matrix equation, and obtains the normal equation matrix R of the TLS-Prony algorithm from the matrix equation; then calculates the approximate parameters, and solves the roots of the characteristic polynomial through the approximate parameters, and finally derives the modal parameters of the identification signal according to the roots of the characteristic polynomial. The method effectively reduces the influence of noise on parameter estimation by simultaneously considering the errors of the observation matrix and the observation vector. In the presence of measurement noise or data error, the TLS-Prony algorithm can more accurately fit the signal model by optimizing the overall least squares solution, thereby improving the accuracy of parameter estimation; in addition, the TLS-Prony algorithm can effectively identify and exclude outliers or noise components in the data through technical means such as singular value decomposition (SVD), thereby avoiding the matrix ill-conditioning problem caused by the traditional prony algorithm. It can be seen that the embodiments of the present invention can solve the detection problem of the strongly coupled, multi-modal, nonlinear, wide-band time-varying oscillation phenomena existing in the "dual-high" power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0066] Figure 1 A schematic flow chart of a method for decomposing modal parameters of a response-driven power system wide-band oscillation is provided in an embodiment of the present invention.

[0067] Figure 2 A partial flow chart of a method for decomposing modal parameters of a response-driven power system wide-band oscillation provided by an embodiment of the present invention.

[0068] Figure 3 A partial flow chart of a method for decomposing modal parameters of a response-driven power system wide-band oscillation provided by an embodiment of the present invention.

[0069] Figure 4 A partial flow chart of a method for decomposing modal parameters of a response-driven power system wide-band oscillation provided by an embodiment of the present invention.

[0070] Figure 5 A partial flow chart of a method for decomposing modal parameters of a response-driven power system wide-band oscillation provided by an embodiment of the present invention.

[0071] Figure 6 A schematic diagram of a grid structure of a test system provided in an embodiment of the present invention.

[0072] Figure 7 Schematic diagram of a reconstructed signal obtained in an embodiment of the present invention.

[0073] Figure 8 A schematic structural diagram of a response-driven power system broadband oscillation modal parameter decomposition device provided in an embodiment of the present invention.

[0074] Fig. 9 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0075] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0076] The information collected in the technical solution of the present invention is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0077] The acquisition, transmission, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0078] It should be noted that in the embodiments of the present invention, certain software, components, models and other existing solutions in the industry may be mentioned, which should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present invention, but it does not mean that the inventor has or will necessarily use the solution.

[0079] The response-driven power system broadband oscillation modal parameter decomposition method provided in the embodiment of the present invention is executed by a subject including but not limited to a computer.

[0080] Figure 1 is a flow chart of a method for decomposing modal parameters of a response-driven power system broadband oscillation provided by an embodiment of the present invention, such as Figure 1 As shown, the response-driven power system broadband oscillation modal parameter decomposition method provided by the embodiment of the present invention includes:

[0081] S100, obtaining a signal related to the power system and the broadband oscillation;

[0082] In step S100, the signals involved in the broadband oscillation include electrical quantities such as voltage, current, active / reactive power, etc. The fluctuations of these signals within a broadband range can be used to monitor and analyze the broadband oscillation phenomenon.

[0083] S200, performing noise reduction processing on the signal related to the broadband oscillation of the power system through an SG filter to obtain a signal to be identified;

[0084] In step S200, in the power system, the SG (Savitzky-Golay) filter can be used to process signals related to broadband oscillations to reduce noise and extract the signal to be identified. The working principle of the SG filter is to slide a window on the data, fit a polynomial to the points in the window at each position, and the center point of the window is replaced by the value estimated by the polynomial fit. This process is repeated for each point in the data, effectively smoothing and reducing noise. The key parameters of the SG filter are the window length and the polynomial order. The window length determines the number of neighboring points used to fit the polynomial at each position, and the order of the polynomial determines the degree of the polynomial used for fitting.

[0085] In power systems, SG filters can be used to reduce the noise of broadband oscillation signals. By selecting the appropriate window size and polynomial order, SG filters can effectively remove noise while retaining signal features such as peak value, width and height. This is crucial for analyzing and identifying broadband oscillation signals in power systems, as these signals may be interfered with by noise, thus affecting the authenticity and reliability of the signal.

[0086] In general, the SG filter can effectively reduce noise and obtain the signal to be identified by performing local polynomial fitting on the broadband oscillation signal in the power system, which is of great significance for the stability analysis and fault diagnosis of the power system.

[0087] S300, establishing a differential equation of the signal to be identified;

[0088] In step S300, the difference equation is a mathematical model describing a discrete time system, which is an equation composed of unknown output sequence terms and input sequence terms. By establishing a difference equation of the signal to be identified, the broadband oscillation signal in the power system can be analyzed and processed.

[0089] S400, expressing the differential equation in a matrix form to obtain a matrix equation;

[0090] In step S400, after establishing the difference equation, the minimum error function between the given identification signal and the approximate signal is obtained; then the TS solution is obtained through the TS algorithm to construct the matrix equation.

[0091] S500, obtaining a normal equation matrix of the TLS-Prony algorithm according to the matrix equation;

[0092] In step S500, the TLS-Prony algorithm is a method for estimating natural frequency from noise data. The TLS-Prony algorithm can be used to extract characteristic parameters from the denoised signal (matrix equation) to obtain the normal equation matrix.

[0093] S600, solving the normal equation matrix to obtain the roots of the characteristic polynomial;

[0094] In step S600, after solving the normal equation matrix, the roots of the characteristic polynomial can be obtained.

[0095] S700: Obtain modal parameters of the signal to be identified according to the roots of the characteristic polynomial.

[0096] In step S700, the modal parameters of the signal to be identified can be derived according to the roots of the characteristic polynomial.

[0097] The response-driven power system broadband oscillation modal parameter decomposition method provided by the embodiment of the present invention first uses the Savizky-Golay filter to achieve data smoothing; then establishes a differential equation to find the minimum error function of a given identification signal and an approximate signal; secondly, obtains the TS solution through the TS algorithm, constructs a matrix equation, and obtains the normal equation matrix R of the TLS-Prony algorithm from the matrix equation; then calculates the approximate parameters, and solves the roots of the characteristic polynomial through the approximate parameters, and finally derives the modal parameters of the identification signal according to the roots of the characteristic polynomial. This method effectively reduces the influence of noise on parameter estimation by simultaneously considering the errors of the observation matrix and the observation vector. In the presence of measurement noise or data error, the TLS-Prony algorithm can more accurately fit the signal model by optimizing the overall least squares solution, thereby improving the accuracy of parameter estimation; in addition, the TLS-Prony algorithm can effectively identify and exclude outliers or noise components in the data through technical means such as singular value decomposition (SVD), thereby avoiding the matrix ill-conditioning problem caused by the traditional prony algorithm. It can be seen that the embodiments of the present invention can solve the detection problem of the strongly coupled, multi-modal, nonlinear, wide-band time-varying oscillation phenomena existing in the "dual-high" power system.

[0098] In some embodiments, the method further includes: determining a broadband oscillation state of the power system according to a modal parameter of the signal to be identified.

[0099] like Figure 2 As shown, in some embodiments, the above step S200 may include:

[0100] S201, filtering the signal using a combination of different window lengths and polynomial orders;

[0101] S202, for each of the combinations, calculating the waveform similarity coefficient and mean square error of the signal before and after filtering;

[0102] S203, determining the parameters of the SG filter based on the waveform similarity coefficient and the mean square error;

[0103] S204 , using the SG filter with the parameters set to perform noise reduction processing on the signal related to the broadband oscillation to obtain a signal to be identified.

[0104] Specifically, a suitable Savizky-Golay filter is designed to perform noise reduction preprocessing on the acquired signal data;

[0105] Try different combinations of window lengths and polynomial orders, and calculate the waveform similarity coefficient NCC and mean square error MSE of the signal before and after signal filtering. The calculation formula is as follows:

[0106]

[0107] Among them, x1(i) and x2(i) are the original signal and the denoised signal respectively; N is the number of decomposed modes; the parameters of the Savizky-Golay filter are determined based on the waveform similarity coefficient NCC and the mean square error MSE.

[0108] The SG filter with the parameters set is used to perform noise reduction processing on the signal related to the broadband oscillation to obtain a signal to be identified.

[0109] like Figure 3 As shown, in some embodiments, the above step S300 may include:

[0110] S301, establishing a fitting function of the signal to be identified;

[0111] S302, establishing a differential equation according to the fitting function;

[0112] S303, establishing an error function between the signal to be identified and the approximate signal obtained by the differential equation;

[0113] S304, optimizing the system parameters of the difference equation by using the TS algorithm with the goal of minimizing the error function to obtain the optimal system parameters;

[0114] S305: Set the system parameters of the differential equation to the optimal system parameters to obtain the differential equation of the signal to be identified.

[0115] Specifically, assuming that the signal to be identified is a linear combination of a series of exponential functions, its fitting function at time n is:

[0116]

[0117] Where: p is the given order.

[0118]

[0119] In the formula, A i ,θ i , α i Respectively represent the modal signal amplitude, phase, and attenuation factor;

[0120] f i represents the modal frequency in Hz; Δt represents the sampling interval.

[0121] The difference equation constructed by mathematical derivation is:

[0122]

[0123] Through least square estimation, in order to minimize the minimum error function, that is, the sum of squares of errors between a given signal and an approximate value, a set of linear matrix equations is obtained and defined as r(i, j).

[0124] like Figure 4 As shown, in some embodiments, the above step S500 may include:

[0125] S501, applying singular value decomposition to the matrix equation to obtain a decomposition result;

[0126] S502: Establish a normal equation matrix of the TLS-Prony algorithm according to the decomposition result.

[0127] Specifically, further expanding the matrix equation r(i, j) can obtain the matrix R of the normal equation of the Prony algorithm, and the form of R is:

[0128]

[0129] In order to reduce the influence of disturbances in the matrix equation on the calculation results, TLS is used instead of the least squares method in the process of solving the matrix equation. The augmented matrix C is defined as [A, b], and the singular value decomposition of the matrix C is obtained:

[0130] C=UΣV H (7)

[0131] Where: U and V represent the left and right singular vector matrices respectively, and both are orthogonal matrices; ∑ represents the diagonal matrix composed of the singular values ​​of the matrix C.

[0132] The matrix V is in the following form:

[0133]

[0134] When v in formula (8) (n+1),(n+1) ≠0, the solution under the TLS algorithm can be obtained as:

[0135]

[0136] like Figure 5 As shown, in some embodiments, the above step S600 may include:

[0137] S601, constructing coefficients of characteristic polynomials according to the normal equation matrix;

[0138] S602: Solve the characteristic polynomial according to the coefficients of the characteristic polynomial to obtain the roots of the characteristic polynomial.

[0139] Specifically,

[0140] Solve equation (6) using the TLS method to determine the effective rank p of the matrix R and the AR parameters α1,…,α p .

[0141] According to the effective rank p of the matrix R and the AR parameters α1,…,α p Construct the coefficients of the characteristic polynomial and obtain the following characteristic polynomial (10):

[0142] 1+α1z -1 +…+α p z -p =0 (10)

[0143] According to the root derivation of the characteristic polynomial in formula (10), the modal parameters of the identified signal are shown in formula (11):

[0144]

[0145] In the formula, A i 、f i , α i , Represent the modal signal amplitude, frequency, attenuation factor, and phase respectively.

[0146] In order to verify that the response-driven power system broadband oscillation modal parameter decomposition method provided in this embodiment is effective in a large system, a 10,000-node electromagnetic transient simulation test system is analyzed below.

[0147] Figure 6 It is a schematic diagram of the grid of the test system, in which the sending-end partition 1S is for the high-proportion renewable energy wind-fire bundled transmission, and the dotted box in the figure indicates the doubly-fed wind farm transmission system.

[0148] The system is set to run in the peak 104 mode, and the relevant control parameters of the 1SWP_156 fan of the 10,000-node full electromagnetic simulation test system under the 104 mode are modified to excite broadband oscillation, and the dominant mode is improved by TLS-Prony solution. The reconstructed signal is as follows: Figure 7 The modal parameters of the oscillation signal are shown in Table 1.

[0149] Table 1 Dominant modal parameters of outlet current components

[0150] Modal Amplitude / kV Attenuation Factor Frequency / Hz Phase IMF1 0.080 0.012 49.957 0.578 IMF2 0.016 -0.518 83.346 -2.896 IMF3 0.006 -342.013 275.520 -0.534

[0151] The waveform similarity coefficient NCC and mean square error MSE are calculated to evaluate the noise reduction effect. The closer the NCC is to 1, the smaller the MSE is, and the better the effect is. As shown in formula (12), the calculated NCC = 0.9972, MSE = 3.7021×10 -5 , meeting the error requirements.

[0152]

[0153] It can be seen that the response-driven power system broadband oscillation modal parameter decomposition method provided by the embodiment of the present invention realizes data smoothing and denoising through the Savizky-Golay filter, which can ensure that the shape and width of the signal remain unchanged and retain the effective characteristics of the signal to the greatest extent. The noise of the wind farm grid-connected system is mostly a high-frequency signal, so the filter can successfully smooth the noise signal, greatly improving the anti-interference ability of the identification scheme; by simultaneously considering the errors of the observation matrix and the observation vector, the influence of noise on parameter estimation is effectively reduced. The TLS-Prony algorithm can fit the signal model more accurately by optimizing the overall least squares solution, thereby improving the accuracy of parameter estimation; in addition, the TLS-Prony algorithm can effectively identify and eliminate outliers or noise components in the data through technical means such as singular value decomposition (SVD), thereby avoiding the matrix pathological problem caused by the traditional prony algorithm.

[0154] Based on the same inventive concept, an embodiment of the present invention also provides a response-driven power system broadband oscillation modal parameter decomposition device, as described in the following embodiments. Since the principle of solving the problem by the device is similar to the response-driven power system broadband oscillation modal parameter decomposition method, the implementation of the device can refer to the implementation of the response-driven power system broadband oscillation modal parameter decomposition method, and the repeated parts will not be repeated.

[0155] like Figure 8 As shown, an embodiment of the present invention provides a response-driven power system wide-band oscillation modal parameter decomposition device, comprising:

[0156] An acquisition module 21 is used to acquire a signal related to the wide-band oscillation of the power system;

[0157] A noise reduction module 22, configured to perform noise reduction processing on the signal related to the broadband oscillation of the power system through an SG filter to obtain a signal to be identified;

[0158] An establishing module 23, used for establishing a differential equation of the signal to be identified;

[0159] A first conversion module 24, used to express the differential equation in a matrix form to obtain a matrix equation;

[0160] A second conversion module 25, used to obtain a normal equation matrix of the TLS-Prony algorithm according to the matrix equation;

[0161] A first solving module 26, used for solving the normal equation matrix to obtain the roots of the characteristic polynomial;

[0162] The second solving module 27 is used to obtain the modal parameters of the signal to be identified according to the roots of the characteristic polynomial.

[0163] The response-driven power system broadband oscillation modal parameter decomposition device provided by the embodiment of the present invention first uses the Savizky-Golay filter to achieve data smoothing; then establishes a differential equation to find the minimum error function of a given identification signal and an approximate signal; secondly, obtains the TS solution through the TS algorithm, constructs a matrix equation, and obtains the normal equation matrix R of the TLS-Prony algorithm from the matrix equation; then calculates the approximate parameters, and solves the roots of the characteristic polynomial through the approximate parameters, and finally derives the modal parameters of the identification signal according to the roots of the characteristic polynomial. This method effectively reduces the influence of noise on parameter estimation by simultaneously considering the errors of the observation matrix and the observation vector. In the presence of measurement noise or data error, the TLS-Prony algorithm can more accurately fit the signal model by optimizing the overall least squares solution, thereby improving the accuracy of parameter estimation; in addition, the TLS-Prony algorithm can effectively identify and exclude outliers or noise components in the data through technical means such as singular value decomposition (SVD), thereby avoiding the matrix ill-conditioning problem caused by the traditional prony algorithm. It can be seen that the embodiments of the present invention can solve the detection problem of the strongly coupled, multi-modal, nonlinear, wide-band time-varying oscillation phenomena existing in the "dual-high" power system.

[0164] In some embodiments, the apparatus further comprises:

[0165] The determination module is used to determine the broadband oscillation state of the power system according to the modal parameters of the signal to be identified.

[0166] In some embodiments, the noise reduction module is specifically used to:

[0167] Filter the signal using a combination of different window lengths and polynomial degrees;

[0168] For each of the combinations, calculating the waveform similarity coefficient and mean square error of the signal before and after filtering;

[0169] Determine the parameters of the SG filter based on the waveform similarity coefficient and the mean square error;

[0170] The SG filter with the parameters set is used to perform noise reduction processing on the signal related to the broadband oscillation to obtain a signal to be identified.

[0171] In some embodiments, the establishment module is specifically used to:

[0172] Establishing a fitting function of the signal to be identified;

[0173] According to the fitting function, a difference equation is established;

[0174] Establishing an error function between the signal to be identified and the approximate signal obtained by the differential equation;

[0175] With the goal of minimizing the error function, the system parameters of the difference equation are optimized by the TS algorithm to obtain the optimal system parameters;

[0176] The system parameters of the differential equation are set to the optimal system parameters to obtain the differential equation of the signal to be identified.

[0177] In some embodiments, the second conversion module is specifically configured to:

[0178] Applying singular value decomposition to the matrix equation to obtain a decomposition result;

[0179] According to the decomposition results, the normal equation matrix of the TLS-Prony algorithm is established.

[0180] In some embodiments, the first solution module is specifically used to:

[0181] According to the normal equation matrix, construct coefficients of characteristic polynomials;

[0182] The characteristic polynomial is solved according to the coefficients of the characteristic polynomial to obtain the roots of the characteristic polynomial.

[0183] Fig. 9 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as Fig. 9 As shown, the electronic device 003 includes a processor 301, a memory 302 and a bus 303. The processor 301 and the memory 302 communicate with each other via the bus 303.

[0184] The processor 301 is used to call the program instructions in the memory 302 to execute the methods provided by the above method embodiments.

[0185] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned response-driven power system wide-band oscillation modal parameter decomposition method is implemented.

[0186] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned response-driven power system wide-band oscillation modal parameter decomposition method is implemented.

[0187] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0188] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0189] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0190] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0191] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for decomposing modal parameters of a response-driven power system broadband oscillation, characterized in that: include: Acquire signals related to broadband oscillations in power systems; Performing noise reduction processing on the signal related to the broadband oscillation of the power system by using an SG filter to obtain a signal to be identified; Establishing a differential equation of the signal to be identified; Expressing the difference equation in matrix form to obtain a matrix equation; Obtaining the normal equation matrix of the TLS-Prony algorithm according to the matrix equation; Solving the normal equation matrix to obtain roots of the characteristic polynomial; According to the roots of the characteristic polynomial, the modal parameters of the signal to be identified are obtained.

2. The method according to claim 1, characterized in that The method further comprises: The broadband oscillation state of the power system is determined according to the modal parameters of the signal to be identified.

3. The method according to claim 1, characterized in that The signal is subjected to noise reduction processing by an SG filter to obtain a signal to be identified, including: Filter the signal using a combination of different window lengths and polynomial degrees; For each of the combinations, calculating the waveform similarity coefficient and mean square error of the signal before and after filtering; Determine the parameters of the SG filter based on the waveform similarity coefficient and the mean square error; The SG filter with the parameters set is used to perform noise reduction processing on the signal related to the broadband oscillation to obtain a signal to be identified.

4. The method according to claim 1, characterized in that: Establishing the differential equation of the signal to be identified includes: Establishing a fitting function of the signal to be identified; According to the fitting function, a difference equation is established; Establishing an error function between the signal to be identified and the approximate signal obtained by the differential equation; With the goal of minimizing the error function, the system parameters of the difference equation are optimized by the TS algorithm to obtain the optimal system parameters; The system parameters of the differential equation are set to the optimal system parameters to obtain the differential equation of the signal to be identified.

5. The method according to claim 1, characterized in that According to the matrix equation, the normal equation matrix of the TLS-Prony algorithm is obtained, including: Applying singular value decomposition to the matrix equation to obtain a decomposition result; According to the decomposition results, the normal equation matrix of the TLS-Prony algorithm is established.

6. The method according to claim 1, characterized in that Solving the normal equation matrix, we can obtain the roots of the characteristic polynomial, including: According to the normal equation matrix, construct coefficients of characteristic polynomials; The characteristic polynomial is solved according to the coefficients of the characteristic polynomial to obtain the roots of the characteristic polynomial.

7. A response-driven power system broadband oscillation modal parameter decomposition device, characterized in that: include: An acquisition module, used for acquiring signals related to the wide-band oscillation of the power system; A noise reduction module, used for performing noise reduction processing on the signal related to the broadband oscillation of the power system through an SG filter to obtain a signal to be identified; An establishing module, used for establishing a differential equation of the signal to be identified; A first conversion module, used for expressing the differential equation in a matrix form to obtain a matrix equation; A second conversion module, used for obtaining a normal equation matrix of the TLS-Prony algorithm according to the matrix equation; A first solving module, used for solving the normal equation matrix to obtain the roots of the characteristic polynomial; The second solving module is used to obtain the modal parameters of the signal to be identified according to the roots of the characteristic polynomial.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images