Forced Oscillation Source Localization Method Based on Wavelet Dissipation Energy Spectrum

Abstract

The present invention discloses a method for locating a forced oscillation source based on a wavelet dissipation energy spectrum, which includes: performing wavelet transform on the electrical variation quantities of a generator to obtain a wavelet coefficient matrix of each electrical variation quantity; calculating the wavelet dissipation energy spectra of each generator at each wavelet coefficient matrix, and positioning the key scale wavelet coefficients strongly related to forced oscillation according to the variation trend of the wavelet dissipation energy spectra at each wavelet scale coefficient; extracting the wavelet dissipation energy spectra corresponding to each generator at the key scale wavelet coefficients; arranging the wavelet dissipation energy spectra of n generators in a columnar manner, and locating the oscillation source according to the proposed criterion for locating a forced oscillation source in a power system based on the wavelet dissipation energy spectrum; and outputting the result of locating the forced oscillation source. The present invention not only inherits the characteristic of the traditional DEF that it can locate the forced oscillation source without a detailed system model and component parameters, but also avoids the need for signal reconstruction processing of the required electrical quantities in the traditional time-domain DEF, thereby improving the efficiency of locating the forced oscillation source.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and in particular, to a method for locating forced oscillation sources in a power system based on a wavelet dissipation energy spectrum (WDES). Background Art

[0002] Forced oscillation (FO) in a power system is usually caused by persistent periodic disturbances. When the disturbance frequency is close to the natural frequency of the system, it will induce a whole-network resonance, seriously threatening the safe and stable operation of the power system [1-3] . On April 21, 2008, a steam turbine in a power plant in Yunnan caused a grid forced oscillation due to improper valve switching operation, resulting in large-scale power oscillations on multiple lines [4] ; on January 11, 2019, a large-scale whole-network forced power oscillation with an oscillation frequency of 0.25 Hz occurred in the eastern power system of the United States due to a fault in the steam turbine control system of a power plant [5] . When a forced oscillation occurs in a power system, if no measures are taken in time to suppress it, it will bring serious consequences to the safe and stable operation of the entire power system. Compared with the natural oscillation of the power system, forced oscillation has the characteristics of strong randomness, fast oscillation start, long duration, and rapid attenuation of the oscillation after eliminating the oscillation source [6] , and there is currently no effective measure to suppress it. The usual treatment measure is to quickly determine the forced oscillation source location (FOSL) and quickly isolate it. Therefore, quickly and accurately locating the oscillation source is the key to suppressing forced oscillation in a power system [7] .

[0003] Currently, the methods for locating forced oscillation sources in a power system are mainly divided into: the method for locating forced oscillation sources based on the dynamic model of the power system [8-10] and the method for locating forced oscillation sources based on wide-area measurement information [11-15] . Although the method for locating forced oscillation sources based on a model can achieve accurate location of the oscillation source, the accuracy of this type of method depends on the accuracy of the established model and parameters. For a complex power system in actual operation, it is usually very difficult to accurately obtain the detailed model and parameters of the system

[16] , so the method for locating forced oscillation sources in a power system based on a model is mostly used for off-line analysis of forced oscillation in a power system. The wide-area measurement system is widely configured in the power system, providing a new idea for locating forced oscillation sources in a power system [11-15]With the help of the wide-area measurement information of the power system, the location method based on the time-domain dissipation energy flow has received attention and development. The dissipation energy flow (DEF) method based on wide-area measurement information is one of them. Its energy consumption is consistent with the damping torque. Then, the wide-area measurement information is directly used to calculate the dissipation energy flow to achieve rapid location of the oscillation source, providing a basis for online location of the oscillation source in the power system.

[13] 。

[0004] However, when using the time-domain dissipation energy flow to locate the forced oscillation source of the power system, it is necessary to transform the wide-area measurement information of the power system from the time domain to the frequency domain, then identify the forced oscillation frequency of the system, and then according to the identified forced oscillation frequency, through band-pass filtering or inverse time-frequency domain transformation, extract the forced oscillation component in the wide-area measurement information, and then calculate the corresponding dissipation energy flow according to the forced oscillation component. [14-15] The calculation process is relatively cumbersome, and the calculation efficiency needs to be improved.

[0005] Therefore, how to get rid of the cumbersome calculation process and quickly and efficiently identify and locate the forced oscillation source of the power system still needs further research.

[0006] References

[0007] [1] Tang Yong, Basic Theory of Forced Power Oscillation in Power System [J]. Power System Technology, 2006, 30(10): 29-33.

[0008] [2] Jiang Tao, Jia Hongjie, Li Guoqing, etc. Power System Homology Identification Based on the Correlation of Wide-Area Measurement Information [J]. Transactions of China Electrotechnical Society, 2017, 32(1): 1-11.

[0009] [3] Chevalier S, Vorobev P, Turitsyn K. A Bayesian Approach to Forced Oscillation Source Location Given Uncertain Generator Parameters [J]. IEEE Transactions on Power Systems, 2019, 34(2): 1641-1649.

[0010] [4] He Yingguang, Liu Tao. Analysis of Low-Frequency Oscillation in Power Grid Caused by Improper Turbine Valve Switching Operation [J]. Electric Power Automation Equipment, 2010, 30(5): 142-145.

[0011] [5] Zhu Lin, Yu Wenpeng, Jiang Zhihao, et al. A comprehensive method to mitigate forced oscillations in large interconnected power Grids[J]. IEEE Access, 2021, 9: 22503-22515.

[0012] [6] Luan Moude. Research on the location of forced oscillation disturbance sources in power systems[D]. Zhejiang University, 2020.

[0013] [7] Wang Maohai, Sun Hao. Online location analysis technology for forced power oscillation sources[J]. Proceedings of the CSEE, 2014, 34(34): 6209-6215.

[0014] [8] Yu Yiping, Min Yong, Chen Lei. Analysis of the steady-state response characteristics of forced power oscillations in multi-machine power systems[J]. Automation of Electric Power Systems, 2009, 33(22): 5-9.

[0015] [9] Yu Yiping, Min Yong, Chen Lei, et al. Analysis of forced power oscillations caused by periodic load disturbances[J]. Automation of Electric Power Systems, 2010, 34(6): 7-11.

[0016]

[10] Dong Chao, Yun Lei, Liu Dichen, et al. Research on the characteristics of forced power oscillations caused by periodic disturbances of prime movers[J]. Power System and Clean Energy, 2012, 28(4): 35-41.

[0017]

[11] Yu Yiping, Min Yong, Chen Lei, et al. Location of forced power oscillation disturbance sources based on energy function[J]. Automation of Electric Power Systems, 2010, 5(5): 1-6.

[0018]

[12] Chen Lei, Min Yong, Hu Wei. An energy-based method for location of power system oscillation source[J]. IEEE Transactions on Power Systems, 2013, 28(2): 828-836.

[0019]

[13] Maslennikov S, Wang Bin, Litvinov E. Dissipating energy flow method for locating the source of sustained oscillations[J]. International journal of electrical power and energy systems, 2017, 88: 55 - 62.

[0020]

[14] Chen Lei, S Ming, Min Yong, et al. Online monitoring of generator damping using dissipation energy flow computed from ambient data[J]. IET Generation Transmission & Distribution, 2017, 11(18): 4430 - 4435.

[0021]

[15] Estevez P G, Marchi P, Galarza C, et al. Non - stationary power system forced oscillation analysis using synchrosqueezing transform[J]. IEEE Transactions on Power Systems, 2020, 36(2): 1583 - 1593.

[0022]

[16] Li Xue, Yu Yang, Jiang Tao, et al. Power system oscillation mode and modal identification method based on sparse enhanced dynamic decoupling[J]. Transactions of China Electrotechnical Society, 2021, 36(13): 2832 - 2843. Summary of the Invention

[0023] The present invention provides a method for locating the forced oscillation source based on the wavelet dissipation energy spectrum. The present invention not only inherits the characteristic of the traditional DEF that it can locate the forced oscillation source without detailed system models and component parameters, but also avoids the need for signal reconstruction processing of the required electrical quantities in the traditional time - domain DEF, significantly improving the location efficiency of the forced oscillation source. Details are described below:

[0024] A method for locating the forced oscillation source based on the wavelet dissipation energy spectrum, the method comprising:

[0025] Perform wavelet transform on the electrical variables of the generator to obtain the wavelet coefficient matrix of each electrical variable;

[0026] Calculate the wavelet dissipation energy spectrum of each generator at each wavelet coefficient matrix, and locate the key scale wavelet coefficients strongly related to forced oscillation according to the change trend of the wavelet dissipation energy spectrum at each wavelet scale coefficient;

[0027] Extract the wavelet dissipation energy spectrum corresponding to each generator at the key scale wavelet coefficients;

[0028] Arrange the wavelet dissipation energy spectra of n generators in a column, and locate the oscillation source according to the proposed criterion for locating the forced oscillation source in the power system based on the wavelet dissipation energy spectrum; output the result of locating the forced oscillation source.

[0029] Among them, the specific calculation of the wavelet dissipation energy spectrum of each generator at each wavelet scale coefficient is as follows:

[0030]

[0031] Among them, is the wavelet dissipation energy of the wavelet scale coefficient c, is the wavelet coefficient of the active power change ΔP ij that is only related to the wavelet scale coefficient c, is the wavelet coefficient of the generator angular velocity change Δω i that is only related to the wavelet scale coefficient c, is the wavelet coefficient of the reactive power change ΔQ ij that is only related to the wavelet scale coefficient c, is the differential of the logarithmic change of the voltage amplitude at node i that is only related to the wavelet scale coefficient c, * is the conjugate, D is the damping coefficient, i is system node i, j is system node j, P is the active power, Q is the reactive power, Δω i is the generator angular velocity change; is the differential of the logarithmic change of the voltage amplitude at node i.

[0032] Furthermore, the specific method of locating the key scale wavelet coefficients strongly related to forced oscillation according to the change trend of the wavelet dissipation energy spectrum at each wavelet scale coefficient is as follows:

[0033]

[0034] In the formula: is the cross-wavelet coefficient of the active power change ΔP ij and the generator angular velocity change Δω i with respect to the scale coefficient c; It is the reactive power variation ΔQ ij and the differential of the logarithmic variation of the voltage amplitude The cross-wavelet coefficient with respect to the scale coefficient c; Re denotes taking the real part.

[0035] Among them, the specific wavelet dissipation energy spectrum corresponding to each generator at the key scale wavelet coefficients is as follows:

[0036] According to the definition of energy spectral density, after presenting the integrand:

[0037]

[0038] In the formula: It is defined as the wavelet dissipation energy spectrum.

[0039] Furthermore, the specific process of locating the forced oscillation source according to the proposed criterion for locating the forced oscillation source of the power system based on the wavelet dissipation energy spectrum is as follows:

[0040] When the wavelet dissipation energy spectrum is used for identifying the forced oscillation mode of the system, the frequency corresponding to the peak value of the wavelet dissipation energy spectrum is the forced oscillation frequency of the system;

[0041] When forced oscillation occurs in the system, if the wavelet dissipation energy spectrum of the generator at the forced oscillation frequency is negative and has the largest absolute value, it is the oscillation source;

[0042] If the wavelet dissipation energy spectrum of the generator at the forced oscillation frequency is negative, but its absolute value is small, it is a non-oscillation source;

[0043] If the wavelet dissipation energy spectrum of the generator at the forced oscillation frequency of the system is positive, it indicates that the generator continuously absorbs energy from the power grid and is a non-oscillation source.

[0044] The beneficial effects of the technical solution provided by the present invention are as follows:

[0045] 1. In the process of realizing the location of the forced oscillation source, the present invention does not need to construct a detailed system energy function model, and can accurately and effectively locate the forced oscillation source based on wide-area measurement data, having strong engineering practicability;

[0046] 2. The present invention converts the traditional time-domain dissipation energy from the time domain to the time-frequency domain through continuous wavelet transform, and then from the time-frequency domain to the frequency domain, so as to analyze the variation law of the dissipation energy in the frequency domain, and further realize the location of the forced oscillation source of the power system;

[0047] 3. When analyzing the wavelet dissipation energy spectrum of each generator, the present invention does not need to reconstruct the time-domain components of each electrical quantity that are strongly correlated with the forced oscillation mode, avoiding complex calculations, improving the calculation speed, and effectively improving the location efficiency of the forced oscillation source. Brief Description of the Drawings

[0048] Figure 1 Flow chart of a method for locating forced oscillation sources in a power system based on wavelet dissipation energy spectrum;

[0049] Figure 2 Topological diagram of the WECC (Western Electricity Coordinating Council)-179 bus system;

[0050] Figure 3 Diagram of electrical parameters at the generator terminals;

[0051] Figure 4 Wavelet dissipation energy spectra of each generator;

[0052] Figure 5 Wavelet dissipation energy spectra of each generator at the forced oscillation frequency. Specific implementation manners

[0053] To make the objectives, technical solutions and advantages of the present invention clearer, the following further describes the implementation manners of the present invention in detail.

[0054] Quickly and accurately locating the forced oscillation sources in a power system is of great significance for preventing power system disconnection and large-scale power outage accidents caused by forced oscillations. However, the current time-domain methods for locating forced oscillation sources have a relatively complex calculation process and the calculation efficiency needs to be improved. For this reason, the embodiments of the present invention propose a method and device for locating forced oscillation sources in a power system based on wavelet dissipation energy spectrum.

[0055] First, taking the wide-area measurement information of the power system as the input, calculate the deviation value of the input data; then, perform wavelet transform on the deviation value of the input data to obtain the corresponding wavelet coefficient matrix; further, calculate the wavelet dissipation energy spectra of each generator, and find out the key wavelet scale coefficients corresponding to the forced oscillation modes of the power system; extract the wavelet dissipation energy spectra corresponding to the key wavelet scale coefficients; and then locate the forced oscillation sources in the power system according to the wavelet dissipation energy spectrum location criterion to achieve the location of the oscillation sources.

[0056] Embodiment 1

[0057] A method for locating forced oscillation sources in a power system based on wavelet dissipation energy spectrum according to an embodiment of the present invention, see Figure 1 , the following further describes the embodiments of the present invention in detail, where 101-109 are the detailed steps of the calculation method:

[0058] 101: The phasor measurement unit PMU in the power grid collects the active power, reactive power signals, terminal voltage signals of the generators, and the bus frequency signals at the generator terminals, and uses them as the input data of the present invention;

[0059] 102: Let \(k = 1\), and calculate the change amounts of the active power, reactive power, terminal voltage, and frequency of the \(k\)-th generator.

[0060] 103: Perform wavelet transform on each electrical change amount to obtain the wavelet coefficient matrix of each electrical change amount.

[0061] 104: Calculate the wavelet dissipation energy spectrum of each generator at each wavelet scale coefficient According to the change trend of the wavelet dissipation energy spectrum at each wavelet scale coefficient, locate the key scale wavelet coefficients strongly related to forced oscillation;

[0062] Among them, the wavelet scale coefficient is obtained from the calculation formula of the wavelet parameters of each generator.

[0063] 105: Extract the wavelet dissipation energy spectrum corresponding to each generator at the key scale coefficient;

[0064] 106: Determine whether \(k\) is equal to \(n\). If so, execute step 107; if not, let \(k=k + 1\) and execute step 102;

[0065] 107: Arrange the dissipation energy spectra corresponding to the key scale coefficients of \(n\) generators in a column;

[0066] Among them, after the column arrangement, it is used as the oscillation source location criterion.

[0067] 108: Locate the oscillation source according to the proposed power system forced oscillation source location criterion based on the wavelet dissipation energy spectrum;

[0068] 109: Output the forced oscillation source location result.

[0069] In summary, through the above steps 101 - 109, the embodiment of the present invention can avoid the signal reconstruction problem of the required electrical quantities, improve the location efficiency of the oscillation source, and thus achieve the fast and accurate location of the power system forced oscillation source.

[0070] Embodiment 2

[0071] The following further introduces the solution in Embodiment 1 in combination with specific calculation formulas and examples. See the following description for details:

[0072] 201: Select the active power \(P\), reactive power \(Q\), frequency \(f\), and voltage amplitude \(U\) collected by the PMU of the generator node in the power grid (technical terms well-known to those skilled in the art, not elaborated here) as the input, calculate the logarithm of the generator voltage amplitude, and use it as the input data of the embodiment of the present invention;

[0073] 202: Calculate the change amounts of each required electrical quantity and perform wavelet transform on them;

[0074] Wavelet transform is a signal processing method in the time-frequency domain, which can effectively process non-stationary measurement signals. The wavelet coefficients after wavelet transform of the measurement signal x(t) are as follows:

[0075]

[0076] In the formula, ψ c,τ (t) is the mother wavelet; c is the wavelet scale coefficient; τ is the wavelet displacement coefficient; is the wavelet coefficient at the scale coefficient c and the displacement coefficient τ; * represents conjugation.

[0077] In the embodiment of the present invention, the complex Morlet wavelet is selected as the mother wavelet. According to the selected mother wavelet, and according to Equation (4), the wavelet coefficient matrix obtained by wavelet transform of the terminal frequency f is:

[0078]

[0079] In the formula, is the wavelet coefficient matrix of the terminal frequency f; is the wavelet coefficient vector at the m-th scale coefficient; is the wavelet coefficient at the m-th scale coefficient and the n-th displacement coefficient.

[0080] In the wavelet coefficient matrix of Equation (2), the scale coefficient c corresponds to the system oscillation frequency, and the displacement coefficient τ corresponds to time. Each row of wavelet coefficients in Equation (2) describes the variation characteristics of system parameters in space with time at a certain oscillation frequency; each column of wavelet coefficients describes the contribution degree of different oscillation frequencies of the system to the oscillation of this parameter at the same time section. The wavelet coefficient matrix of Equation (2) describes the response characteristics of system parameters when the system oscillates from the perspective of time and space, and provides a certain reference for revealing the mechanism of system forced oscillation and locating the forced oscillation source with the help of relevant parameters.

[0081] 203: Location of forced oscillation source in power system based on wavelet dissipation energy spectrum;

[0082] According to the theory of dissipation energy flow, the oscillation energy flow from node i to node j is defined as:

[0083]

[0084] In the formula, ΔP ij and ΔQ ij are the changes in active power and reactive power transmitted from node i to node j respectively; Δω i is the change in generator angular velocity; is the differential of the logarithmic change in voltage amplitude at node i.

[0085] According to the wavelet transform theory, it is known that its multiplication formula satisfies:

[0086]

[0087] In the formula, C φ is the wavelet transform admissibility constant; and are the wavelet coefficient matrices of the active power change ΔP ij and the generator angular velocity change Δω i respectively; * represents the conjugate.

[0088] According to equation (4), further transform equation (3) to get:

[0089]

[0090] In the formula, is the time-frequency domain dissipation energy; and are the wavelet coefficients of the reactive power change ΔQ ij and the differential of the logarithm of the voltage amplitude change respectively; * represents the conjugate.

[0091] After integrating equation (5) with respect to the displacement coefficient τ, the frequency domain energy expression for the scale coefficient c can be obtained:

[0092]

[0093] In the formula, is the wavelet dissipation energy of the wavelet scale coefficient c.

[0094] Since the actual measured signal is generally a real signal, the conjugate of the real signal remains unchanged. At this time, according to Parseval's theorem, for two different measured signals ΔP ij (t) and Δω ij (t) satisfy:

[0095]

[0096] In the formula, and are the Fourier transforms of ΔP ij (t) and Δω ij (t) respectively; * represents the conjugate.

[0097] Further transform equation (7) to get:

[0098]

[0099] In the formula, is for ΔP ij (t) and Δωi The cross - energy spectral density of (t); Re represents taking the real part.

[0100] According to Equation (8), further transform Equation (6) to obtain:

[0101]

[0102] Where: is the change in active power ΔP ij and the change in generator angular velocity Δω i The cross - wavelet coefficient with respect to the scale coefficient c; is the change in reactive power ΔQ ij and the differential of the change in the logarithm of the voltage amplitude The cross - wavelet coefficient with respect to the scale coefficient c.

[0103] Further process Equation (9) to obtain:

[0104]

[0105] According to the definition of energy spectral density, extract the integrand in Equation (10):

[0106]

[0107] Equation (11) fully reflects the change of dissipated energy under each wavelet scale coefficient. In the embodiment of the present invention, is defined as the wavelet dissipated energy spectrum.

[0108] Due to the corresponding relationship between the wavelet scale coefficient and the oscillation frequency, the wavelet dissipated energy spectrum calculated by Equation (11) can truly reflect the change trend of the wavelet dissipated energy in the frequency domain, and then the dominant oscillation frequency of the system can be directly determined. When there is an obvious jump in the wavelet dissipated energy spectrum curve, it indicates that there is a certain dominant oscillation mode in the system, and the peak value of the wavelet dissipated energy spectrum corresponds to the oscillation frequency of this dominant oscillation mode. Similarly, when the wavelet dissipated energy spectrum is used for identifying the forced oscillation mode of the system, the frequency corresponding to the peak value of the wavelet dissipated energy spectrum is the forced oscillation frequency of the system.

[0109] When the system undergoes forced oscillation, if the wavelet dissipated energy spectrum of the generator at the forced oscillation frequency is negative and has the largest absolute value, it indicates that the generator continuously injects energy into the power grid, and due to the largest absolute value of the energy, its influence on the forced oscillation is the greatest, and it is the oscillation source; if the wavelet dissipated energy spectrum of the generator at the forced oscillation frequency is negative, but its absolute value is small, it means that the generator has a small influence on this forced oscillation and is a non - oscillation source; if the wavelet dissipated energy spectrum of the generator at the system forced oscillation frequency is positive, it indicates that the generator continuously absorbs energy from the power grid and is a non - oscillation source.

[0110] Embodiment 3

[0111] Combined with specific examples below, for the forced oscillation source localization method based on wavelet dissipation energy spectrum proposed in the embodiments of the present invention, this example takes the WECC 179 - bus system as an example for simulation analysis and verification. The topology diagram of the WECC 179 - bus system is as Figure 2 shown. Set generator G36 as the reference generator, and continuously inject a sine signal with an oscillation frequency of 0.46 Hz into the excitation system of generator G79 as the forced oscillation disturbance signal, with a duration of 20 s. The electrical parameters at the generator terminals during the disturbance are as Figure 3 shown.

[0112] According to the forced oscillation source localization method proposed in the embodiments of the present invention, first calculate Figure 3 the electrical quantity parameters such as the change in active power, change in reactive power, logarithmic change in voltage amplitude, and change in frequency of each generator in, perform wavelet transform on the above - mentioned electrical quantity change parameters to obtain the wavelet coefficient matrix of each electrical quantity of each generator. Then calculate the wavelet dissipation energy spectrum of each generator. The results are as Figure 4 shown.

[0113] Obviously, from Figure 4 it can be seen that: the wavelet dissipation energy spectra of each generator have obvious peaks at c k = 192, and the absolute values of the wavelet dissipation energy of each generator are the largest at this point, indicating that c k = 192 is the key wavelet scale coefficient, corresponding to a dominant oscillation mode of the system. Further, the oscillation frequency and damping ratio of the dominant oscillation mode of the system corresponding to c k = 192 are calculated to be 0.4685 Hz and 0.27% respectively. This dominant oscillation frequency is basically consistent with the injected disturbance frequency of 0.46 Hz, verifying that the method of determining the forced oscillation mode based on the jump of the wavelet dissipation energy spectrum is accurate.

[0114] To further localize the forced oscillation source with an oscillation frequency of 0.4685 Hz, according to the method proposed in the embodiments of the present invention, extract the wavelet dissipation energy spectra of each generator at ck = 192. The results are as Figure 5 shown. Obviously, from Figure 5It can be seen that the wavelet dissipation energy spectrum of the generator G79 at the forced oscillation frequency is negative and has the largest absolute value. According to the criterion for locating the forced oscillation source based on the wavelet dissipation energy spectrum, the generator G79 is the forced oscillation source with an oscillation frequency of 0.4685 Hz. In the figure, the wavelet dissipation energy of some generators at the forced oscillation frequency is negative, but its absolute value is much smaller than that of G79. Although these generators contribute to the forced oscillation, the contribution is very small and can be ignored. In addition, the wavelet dissipation energy of some generators at the forced oscillation frequency of 0.4685 Hz is positive, indicating that these generators continuously consume energy from the power grid and are non-oscillation sources.

[0115] Table 1 Comparison of calculation efficiencies of different methods

[0116]

[0117] Table 1 further compares the calculation time of the method proposed in the embodiment of the present invention with the traditional time-domain dissipation energy flow method and the time-frequency domain dissipation energy flow method in the above two scenarios (the hardware configuration of the calculation platform is: CPU Intel Core i7-9750H, main frequency 2.6 GHz, memory: 16 GB). Obviously, from the results in Table 1, it can be seen that in the single oscillation source location of the WECC179-bus test system, the calculation efficiency of the proposed method is increased by 32.78% and 31.39% respectively compared with the traditional time-domain dissipation energy flow method and the time-frequency domain dissipation energy flow method.

[0118] In summary, through the analysis of the results of the above figures and tables, the results verify that the location result of the wavelet dissipation energy spectrum location oscillation source method proposed by the present method fully conforms to the fault setting, effectively improves the oscillation source location efficiency, and verifies the accuracy and effectiveness of the proposed method.

[0119] In the embodiments of the present invention, except for those with special descriptions for the models of each device, the models of other devices are not limited, as long as the devices can perform the above functions.

[0120] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred embodiment, and the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments.

[0121] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for locating forced oscillation sources based on wavelet dissipation energy spectrum, characterized in that, the method includes: Performing wavelet transform on the electrical change quantities of the generator to obtain the wavelet coefficient matrix of each electrical change quantity; Calculating the wavelet dissipation energy spectrum of each generator at each wavelet scale coefficient, and locating the key scale wavelet coefficients strongly related to forced oscillation according to the change trend of the wavelet dissipation energy spectrum at each wavelet scale coefficient; Extracting the wavelet dissipation energy spectrum corresponding to each generator at the key scale wavelet coefficients; Arranging the wavelet dissipation energy spectra of n generators in a column, and locating the oscillation source according to the proposed criterion for locating forced oscillation sources in power systems based on wavelet dissipation energy spectrum; outputting the location result of the forced oscillation source; wherein, the specific calculation of the wavelet dissipation energy of each generator at each wavelet scale coefficient is: Among them, is the wavelet dissipation energy of the wavelet scale coefficient c, is the active power change ΔP that only relates to the wavelet scale coefficient c ij 's wavelet coefficient, is the generator angular velocity change Δω that only relates to the wavelet scale coefficient c i 's wavelet coefficient, is the reactive power change ΔQ that only relates to the wavelet scale coefficient c ij 's wavelet coefficient, is the differential of the logarithmic change of the voltage amplitude at node i that only relates to the wavelet scale coefficient c 's wavelet coefficient, * is the conjugate, D is the damping coefficient, i is system node i, j is system node j, P is the active power, Q is the reactive power, Δω i is the generator angular velocity change; is the differential of the logarithmic change of the voltage amplitude at node i; The specific wavelet dissipation energy used for locating the key scale wavelet coefficients strongly related to forced oscillation according to the change trend of the wavelet dissipation energy spectrum at each wavelet scale coefficient is: In the formula: is the change in active power ΔP ij and the change in generator angular velocity Δω i are the cross-wavelet coefficients with respect to the scale coefficient c; is the change in reactive power ΔQ ij and the differential of the logarithmic change in voltage amplitude are the cross-wavelet coefficients with respect to the scale coefficient c; Re represents taking the real part; The specific extraction of the wavelet dissipation energy spectrum corresponding to each generator at the key scale wavelet coefficients is: According to the definition of energy spectral density, after presenting the integrand: In the formula: is defined as the wavelet dissipation energy spectrum.

2. The method for locating forced oscillation sources based on wavelet dissipation energy spectrum according to claim 1, characterized in that, the specific location of the oscillation source according to the proposed criterion for locating forced oscillation sources in power systems based on wavelet dissipation energy spectrum is: When the wavelet dissipation energy spectrum is used for identifying the forced oscillation mode of the system, the frequency corresponding to the peak value of the wavelet dissipation energy spectrum is the forced oscillation frequency of the system; When forced oscillation occurs in the system, if the wavelet dissipation energy spectrum of the generator at the forced oscillation frequency is negative and has the largest absolute value, it is the oscillation source; If the wavelet dissipation energy spectrum of the generator at the forced oscillation frequency is negative, but its absolute value is small, it is a non-oscillation source; If the wavelet dissipation energy spectrum of the generator at the forced oscillation frequency of the system is positive, it indicates that the generator continuously absorbs energy from the power grid and is a non-oscillation source.

Images