A novel dynamic fusion and identification method for oscillation multimodal information in wind power generation systems

Abstract

This invention relates to the technical field of power systems and discloses a novel dynamic fusion identification method for multi-mode oscillation information in wind power generation systems. First, frequency analysis is performed on the collected data from the wind power system to obtain the frequency domain characteristics corresponding to the oscillation components at each sampling frequency point. Then, based on the attenuation factor in the frequency domain characteristics, non-power frequency oscillation components are identified. Next, the modal energy and damping ratio corresponding to each non-power frequency oscillation component are calculated. Finally, by comprehensively ranking the modal energy and damping ratio, the non-power frequency oscillation component with the highest ranking is selected as the dominant oscillation mode. Oscillation mode identification is then performed using the sampling frequency point corresponding to the dominant oscillation mode. Compared with existing methods, the identification method of this invention has a fast computation speed, good real-time performance, and wide applicability.

Description

Technical Field

[0001] This invention relates to the technical field of power systems, and more specifically, to a novel dynamic fusion and identification method for oscillation multimodal information in wind power generation systems. Background Technology

[0002] In recent years, with the application of high-proportion power electronic equipment in high-proportion wind power generation, the power system has exhibited a new characteristic of "high power consumption and high efficiency." In light of this, the "mechanical-electrical-magnetic" interactions between power electronic devices and between them and the power grid can cause periodic fluctuations in voltage, current, and power information over time, easily leading to new types of oscillations with large-scale frequency variations.

[0003] The occurrence of this new type of oscillation is closely related to information such as wind speed, number of wind turbines, and system impedance. Furthermore, oscillations of different frequencies can occur in different wind farms or even within the same wind farm due to variations in system operating parameters. To effectively suppress these oscillations, it is essential to first identify the oscillation frequency using effective methods.

[0004] In invention patent CN106300345, an improved method for identifying low-frequency oscillation parameters using the Prony algorithm is proposed. This method primarily employs neural networks and gradient descent to adjust weights to achieve the Prony algorithm's solution, but it only targets the detection of low-frequency oscillations in power systems. In invention patent CN104852392A, a method for calculating the attenuation coefficient of subsynchronous oscillation modes based on the Prony algorithm is proposed, but it focuses on the torsional vibration modes and subsynchronous oscillation frequencies of turbine generator shaft systems. However, the "wideband" characteristic of novel oscillations in wind power systems is manifested not only in subsynchronous oscillation frequencies but also in supersynchronous oscillation frequencies. Therefore, how to effectively identify oscillation frequencies within a sufficiently "wideband" range is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides a novel dynamic fusion and identification method for multi-mode oscillation information in wind power generation systems, which solves the technical problem that existing methods cannot identify subsynchronous and supersynchronous oscillations in wind power systems.

[0006] This invention can be achieved through the following technical solutions:

[0007] A novel dynamic fusion and identification method for oscillation multimode information in wind power generation systems is proposed. First, frequency analysis is performed on the data collected from the wind power system to obtain the frequency domain characteristics of the oscillation components at each sampling frequency point. Then, based on the attenuation factor in the frequency domain characteristics, non-power frequency oscillation components are identified. Next, the modal energy and damping ratio corresponding to each non-power frequency oscillation component are calculated. Finally, by comprehensively ranking the modal energy and damping ratio, the non-power frequency oscillation component with the highest ranking is selected as the dominant oscillation mode. The sampling frequency point corresponding to the dominant oscillation mode is used for oscillation mode identification.

[0008] Furthermore, the oscillation modes include low-frequency oscillation mode LFO, sub / supersynchronous oscillation mode SSO, and medium-high frequency oscillation mode MHFO;

[0009] If the sampling frequency point corresponding to the dominant oscillation mode is between 0 and 10 Hz, it is identified as a low-frequency oscillation mode (LFO); if the sampling frequency point corresponding to the dominant oscillation mode is between 10 and 100 Hz, it is identified as a subsynchronous / supersynchronous oscillation mode (SSO); if the sampling frequency point corresponding to the dominant oscillation mode is between 100 and several kHz, it is identified as a mid-to-high frequency oscillation mode (MHFO).

[0010] Furthermore, when using a multimodal information dynamic fusion and identification method to perform frequency domain analysis on the sampled data, the time domain expression of the sampled data is denoted as:

[0011]

[0012] In the formula, p is the model order, fi is the frequency, Ai is the amplitude, and θ is the amplitude. i For phase, σ i It is the attenuation factor;

[0013] Furthermore, if the attenuation factor corresponding to the oscillation component is less than zero, then the corresponding oscillation component is a non-power frequency oscillation component.

[0014] Furthermore, using the following formula, the modal energy and damping ratio of each non-power frequency oscillation component are calculated. The modal energy is given the primary priority and the low damping ratio is given the secondary priority. The bubble sort method is used to sort the components, and the non-power frequency oscillation component ranked first is selected as the dominant oscillation mode.

[0015] The modal energy E i and damping ratio ξ i The mathematical expression for can be represented as follows.

[0016]

[0017]

[0018] In the formula, A i For amplitude, z irepresents the matrix coefficients in the discrete mathematical model after discretization, M represents the number of sampling frequency points, and i represents the number of the sampling frequency points.

[0019] The beneficial technical effects of this invention are as follows:

[0020] This invention can identify the oscillation frequency, amplitude, phase, and attenuation factor in voltage / current signals for the specific operating condition of subsynchronous / supersynchronous oscillations in high-proportion wind power generation systems. It then filters out non-power frequency oscillation components with attenuation factors less than zero and their information. The damping ratio and modal energy of the filtered non-power frequency oscillation components are autonomously calculated. Based on the principle of prioritizing high modal energy and secondary prioritizing low damping ratio, the non-power frequency oscillation components are sorted and the corresponding modal information is stored. Finally, the dominant oscillation mode is determined based on the sorting results. This invention enables wide-area information identification of wind power generation oscillations, with an identification frequency range reaching several kHz. It can identify both subsynchronous and supersynchronous oscillations and further support oscillation suppression. Compared to existing methods, the identification method of this invention has a faster calculation speed, better real-time performance, and wider applicability. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0022] Figure 2 This is a schematic diagram illustrating the application of the novel dynamic fusion and identification method for oscillation multimodal information in wind power generation systems according to the present invention in a wind power system.

[0023] Figure 3 This diagram illustrates the comparison between the collected current signal discretization processing and the fitting simulation results and the actual measurement results using the novel dynamic fusion identification method for oscillation multimodal information of wind power generation system according to the present invention. The red line represents the fitted simulation results, and the blue line represents the actual measurement results.

[0024] Figure 4 The illustration shows the results of a specific embodiment of the novel dynamic fusion identification method for oscillation multimodal information of wind power generation system according to the present invention. Detailed Implementation

[0025] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0026] like Figure 1As shown, this invention provides a novel dynamic fusion and identification method for oscillation multimodal information in wind power generation systems. First, frequency analysis is performed on the collected data from the wind power system to obtain the frequency domain characteristics corresponding to the oscillation components at each sampling frequency point. Then, based on the attenuation factor in the frequency domain characteristics, non-power frequency oscillation components are identified. Next, the modal energy and damping ratio corresponding to each non-power frequency oscillation component are calculated. Finally, by comprehensively ranking the modal energy and damping ratio, the non-power frequency oscillation component ranked first is selected as the dominant oscillation mode, and the sampling frequency point corresponding to the dominant oscillation mode is used for oscillation mode identification.

[0027] The specific implementation method is as follows:

[0028] Step 1: Identification of Oscillation Frequency Signal

[0029] The time-domain expression of the oscillating signal x'(t) can be expressed as:

[0030]

[0031] In the formula, p is the model order, and f i Let i be the frequency, and i be the number of the fitted exponential function, which can represent 1, 2, 3, ..., A. i θ represents the amplitude, t represents time, and θ represents the time. i For phase, σ i This is the attenuation factor.

[0032] Discretizing the oscillating signal x(t), the resulting discrete mathematical model at each sampling time can be expressed as:

[0033]

[0034] Where zi and bi are matrix coefficients; Δt is the sampling time interval; and N is the number of data sampling points, N≥2p.

[0035] Further calculations are performed on the coefficients of the discrete mathematical model. Based on this, the frequency domain information of the obtained sampling frequency points is solved, including information such as the frequency, amplitude, phase, and attenuation factor of the sampling frequency points. The specific expression is as follows:

[0036] s i =In|Z i | / △t (5)

[0037] f i =arg[Im(Z i ) / Re(Z i )] / 2p△t (6)

[0038] A i =b i | (7)

[0039] θ i =arg[Im(b i ) / Re(b i (8)

[0040] Step 2: Identification of Non-Power Frequency Oscillation Modes

[0041] The attenuation factor at the corresponding sampling frequency point obtained by the multimodal information fusion identification method is extracted. The attenuation factor reflects the attenuation trend and degree of the system at that frequency point. The stability of the system at each sampling frequency point is analyzed based on the attenuation factor. When the attenuation factor σ at the sampling frequency point... i When the value is greater than zero, the system is highly likely to oscillate at that frequency. Extract and store the attenuation factor σ. i Dynamic frequency domain information of non-power frequency sampling points with a value greater than zero.

[0042] Step 3: Identification of the dominant oscillation mode

[0043] Determining the attenuation factor σ i Based on the non-power frequency oscillation components that are less than zero, the modal energy and damping ratio are autonomously calculated by combining their dynamic frequency domain information, and then the dominant oscillation mode in the given oscillation signal is determined.

[0044] Oscillation mode is all the information that characterizes the oscillation frequency, including oscillation frequency, amplitude, phase, energy, damping ratio, attenuation factor, etc. The most dominant oscillation mode is identified by two indicators: 1) modal energy; 2) damping ratio. The modal energy and damping ratio corresponding to each non-power frequency oscillation component can be calculated using the following formula.

[0045] The mathematical expressions for modal energy and damping ratio can be represented as follows:

[0046]

[0047] In the formula, A i denoted as amplitude, M as the number of sampling frequency points, and i as the sampling frequency point number.

[0048] Modal energy characterization system stability at the non-power frequency sampling point, when A i and z i When the value is large, the modal energy E i A larger damping ratio ξ results in more severe oscillations, and vice versa. i The damping ratio characterizes the rate of oscillation decay. The larger the value of the damping ratio, the faster the oscillation amplitude decays, and the smaller the damping ratio, the slower the decay rate. Therefore, both can better characterize the dynamic change process of the system at that frequency point.

[0049] Based on the internally calculated modal energy and damping ratio results mentioned above, modal energy is prioritized first, and low damping ratio is secondary, using a bubble sort method to sort the various non-power frequency oscillation components. The use of two criteria is to avoid identical modal energies among the non-power frequency oscillation components at the sampling frequency point; furthermore, damping ratio can be used to determine the dominant oscillation mode.

[0050] The autonomously calculated modal energies and damping ratios at sampling frequency points are sorted, prioritizing modal energy (higher energy, higher position) and vice versa. When modal energies at different sampling frequencies are found to be consistent, damping ratio is further considered as a sorting rule: lower damping ratio, higher position, and vice versa. A bubble sort method is used to sort non-power frequency oscillation components in descending order, obtaining descending oscillation frequencies, and simultaneously storing the frequency domain information of the corresponding oscillation frequencies.

[0051] Step 4: Oscillation Mode Identification

[0052] From the descending oscillation frequency sequence obtained in step three, the non-power frequency oscillation component that ranks first is selected as the dominant oscillation mode.

[0053] Oscillation mode positioning

[0054] The oscillation patterns obtained after bubble sort are located by their descending order. The location object is the oscillation frequency, and the location principle is the specified frequency range, which is 0-10Hz, 10-100Hz, and 100-several kHz.

[0055] Oscillation mode classification

[0056] The obtained oscillation mode localization results are classified into the following categories: 0-10Hz is low frequency oscillation mode (LFO), 10-100Hz is sub / sup-synchronous oscillation mode (SSO), and 100-several kHz is medium to high frequency oscillation mode (MHFO).

[0057] To verify the feasibility of the novel dynamic fusion and identification method for oscillation multimodal information in wind power generation systems, we conducted the following experiments:

[0058] exist Figure 2In the wind power grid-connected system shown, the grid frequency is 50Hz. When there is no oscillation, the system only has the power frequency signal, i.e., 50Hz. As the number of wind turbines increases, non-power frequency signals, i.e., 22Hz and 78Hz, appear in the system. The novel dynamic fusion identification method for oscillation multimodal information of the wind power system of this invention is used to collect the current signal of the wind power grid-connected system. The sampling frequency interval is 2Hz, and the signal is discretized. The fitting result is as follows: Figure 3 As shown in the figure, the fitting results are basically consistent with the measurement results.

[0059] Furthermore, by collecting frequency information and using modules such as oscillation energy and damping ratio calculations, the dominant oscillation mode was determined, and the oscillation mode was finally identified. Analysis of the data in Table 1 shows that the attenuation factor for 22Hz is 0.11, and the attenuation factor for 78Hz is 0.12. Therefore, it is determined that there is a subsynchronous oscillation frequency of 22Hz and a supersynchronous oscillation frequency of 78Hz. Additionally, the energy of 22Hz is 7.9, and the energy of 78Hz is 2. Since energy is the dominant factor at this point, 22Hz can be determined as the dominant oscillation frequency. Specific results are as follows: Figure 4 As shown in the figure. Therefore, the feasibility of the proposed dynamic fusion and identification method for oscillatory multimodal information is further verified.

[0060] Table 1 Calculation results of modal energy and damping ratio

[0061]

[0062] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples. Various changes or modifications can be made to these embodiments without departing from the principles and essence of the present invention. Therefore, the scope of protection of the present invention is defined by the appended claims.

Claims

1. A novel dynamic fusion and identification method for oscillation multimodal information in a wind power generation system, characterized in that: First, frequency analysis is performed on the data collected from the wind power system to obtain the frequency domain characteristics of the oscillation components at each sampling frequency point. Then, based on the attenuation factor in the frequency domain characteristics, the non-power frequency oscillation components are identified. Next, the modal energy and damping ratio corresponding to each non-power frequency oscillation component are calculated. Finally, by comprehensively ranking the modal energy and damping ratio, the non-power frequency oscillation component ranked first is selected as the dominant oscillation mode. The oscillation mode is identified based on the sampling frequency point corresponding to the dominant oscillation mode. Using the following formula, calculate the modal energy and damping ratio of each non-power frequency oscillation component. With modal energy as the primary priority and low damping ratio as the secondary priority, sort them using the bubble sort method, and select the non-power frequency oscillation component ranked first as the dominant oscillation mode. The modal energy and damping ratio The mathematical expression for it is as follows: In the formula, A i For amplitude, z i Here, M represents the matrix coefficients in the discretized mathematical model, i represents the sampling frequency points, and f represents the sampling frequency point number. i For frequency, σ i This is the attenuation factor.

2. The novel dynamic fusion and identification method for oscillation multimodal information of wind power generation system according to claim 1, characterized in that: The oscillation modes include low-frequency oscillation mode LFO, sub / supersynchronous oscillation mode SSO, and medium-high frequency oscillation mode MHFO; If the sampling frequency point corresponding to the dominant oscillation mode is between 0 and 10 Hz, it is identified as a low-frequency oscillation mode (LFO); if the sampling frequency point corresponding to the dominant oscillation mode is between 10 and 100 Hz, it is identified as a subsynchronous / supersynchronous oscillation mode (SSO); if the sampling frequency point corresponding to the dominant oscillation mode is between 100 Hz and several kHz, it is identified as a mid-to-high frequency oscillation mode (MHFO).

3. The novel dynamic fusion and identification method for oscillation multimodal information of wind power generation system according to claim 1, characterized in that: When using a multimodal information dynamic fusion and identification method to perform frequency domain analysis on the sampled data, Let the time-domain expression of the sampled data be: In the formula, p is the model order, and f i For frequency, A i For amplitude, θ i For phase, σ i This is the attenuation factor.

4. The novel dynamic fusion and identification method for oscillation multimodal information of wind power generation system according to claim 3, characterized in that: If the attenuation factor corresponding to the oscillation component is less than zero, then the corresponding oscillation component is a non-power frequency oscillation component.

Images