A method for wide frequency oscillation protection of new energy power system

Abstract

A method for wide frequency oscillation protection of new energy power system is proposed, which uses real-time measurement of three-phase voltage / current of new energy station sending line. Firstly, the system stability is evaluated by the proposed real-time frequency detection method; secondly, when the system has persistent wide frequency oscillation, the oscillation source is located, and the output of each line to the oscillation is sorted according to the calculation results of the proposed method, and then the station which should be cut off first is found. The characteristics of the method are: using the sampling information in the rolling time window for real-time monitoring and protection; using the frequency measurement value as the stability criterion based on the characteristic system algorithm; when the oscillation occurs, the oscillation frequency mode admittance is analyzed on the dq coordinate axis, so as to quantitatively represent the output of each line to the oscillation.

Description

Technical Field

[0001] This invention belongs to the field of novel power system monitoring and protection technology, specifically involving broadband oscillation stability assessment based on feature realization algorithm, and novel power system broadband oscillation source tracing and new energy wind farm selection and sorting algorithm based on analytical admittance in dq rotating coordinate system. Background Technology

[0002] With the large-scale grid connection of new energy sources, the formation of high-voltage direct current transmission networks, and the commissioning of power electronic loads, modern power systems are facing an increasing number of broadband oscillation problems. In the past decade, numerous broadband oscillation cases have occurred worldwide. Broadband oscillations can damage power equipment, lead to the shutdown of new energy generating units, and seriously affect equipment safety and threaten the stable operation of the system, becoming one of the most serious risks to the stable operation of systems with a high proportion of new energy. The structure of power electronic equipment in new power systems is complex, exhibiting nonlinearity and time-varying characteristics. After broadband oscillations occur, common research methods, such as establishing analytical models and building simulation platforms to analyze the problem, require analytical analysis after the oscillation occurs, lacking real-time capability. Therefore, effective monitoring and control should be implemented immediately upon the occurrence of oscillations to bring them under control as quickly as possible with minimal cost, ensuring the reliable operation of the power system.

[0003] It is foreseeable that broadband oscillation monitoring and control research will become a key focus in the future development of new energy systems. Current wide-area monitoring systems (WAMS) are based on synchronous vector monitoring units (PMUs) and can achieve real-time online monitoring, analysis, and control of large-span power systems. However, existing PMUs and WAMSs are insufficient to meet the needs of centralized broadband, multi-mode oscillation monitoring and protection for new energy aggregation systems. The following aspects require improvement:

[0004] Existing monitoring devices cannot cover the monitoring and protection of low-frequency oscillations from around 0.1Hz to high-frequency oscillations above 1000Hz;

[0005] The reliability of stability assessment for new power systems is not high;

[0006] When a broadband oscillation problem caused by multiple renewable energy power plants is discovered, it is not possible to properly sort the oscillation sources to effectively cut off some power plants and suppress the oscillation.

[0007] Based on the above considerations, regional power grid operation and control personnel urgently need a broadband oscillation monitoring and protection mechanism suitable for high-proportion renewable energy aggregation systems to maintain the reliable and stable operation of renewable energy systems. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a broadband oscillation protection method for new energy power systems. The method involves real-time acquisition of three-phase voltage and current at the transmitting end of new energy power plants, simultaneous calculation of the positive-sequence frequency of the three-phase voltage, and using this frequency to transform the voltage / current from abc stationary coordinate system to a dq rotating coordinate system. If the oscillation judgment condition is met, the oscillation frequency and damping ratio are calculated. The admittance at the transmitting end of the new energy power plant at this oscillation frequency is used to determine whether it is an oscillation source. The admittances are then ranked, and the most severely affected new energy power plants are identified and disconnected. This method avoids analyzing two frequency components caused by coupling of a single oscillation mode, balancing the complexity and effectiveness of priority ranking during new energy power plant disconnection.

[0009] The present invention adopts the following technical solution.

[0010] A method for broadband oscillation protection in new energy power systems, characterized by comprising the following steps:

[0011] Step 1: Use a measuring device to collect the three-phase voltage and current at the output end of any new energy power station line in real time.

[0012] Step 2: Calculate the positive sequence frequency of the three-phase voltage synchronously using DDSRF-PLL to obtain the voltage and current in the dq rotating coordinate system;

[0013] Step 3: In the real-time scrolling window, determine the stability of the positive sequence frequency within the time period: If the oscillation determination condition is not met, return to Step 2 to continue monitoring; if the oscillation determination condition is met, calculate the oscillation frequency and damping ratio, and continue to Step 4 for analysis.

[0014] Step 4: Calculate the admittance of the corresponding oscillation mode on the dq axis using the voltage and current on the dq axis;

[0015] Step 5: Determine whether each new energy power station is an oscillation source by the admittance of its transmission end at the oscillation frequency, sort the admittances, identify the most serious new energy power station, and cut it off.

[0016] Preferably, in step 3, the stability determination is based on the time window with a total duration of T0, where T is the stability value. s The sampling frequency is the positive sequence frequency ω within the real-time time window. + Perform stability assessment;

[0017] The stability determination method combines two factors: first, it determines the frequency ω within the time window. + Does the difference between local extrema exceed the warning value ±Δf? lim (Hz), and the number of occurrences exceeds N p If the conditions are met, the feature system implementation algorithm is activated to calculate the positive sequence frequency ω. +The oscillation mode frequencies (f1, f2, ..., f) in N ) and damping ratio (D1,D2,…,D N N represents the number of monitored oscillation modes; stability is determined based on the oscillation mode damping ratio.

[0018] In any f n Under the corresponding mode n, if the damping ratio satisfies 0% <D n <D lim If the system is considered to be at critical stability with a risk of oscillation, an early warning can be issued, and the power generation station can be temporarily not shut down while monitoring continues; if the oscillation mode damping ratio satisfies D... n If the percentage is ≤0%, it indicates that the system has experienced wideband oscillations, triggering an early warning and preparing for the next step of disconnecting new energy power plants.

[0019] Preferably, in step 4, the voltage and current on the dq axis are used to calculate the admittance of the corresponding oscillation mode, and the corresponding oscillation mode f is calculated on the dq axis. n The voltage / current vector under the current, i.e., V d (f n V q (f n ),I d (f n ) and I q (f n And calculate the admittance matrix on the dq axis according to the following formula:

[0020]

[0021] Where m = 1, 2, 3, ..., M, M is the total number of new energy power stations measured; n = 1, 2, 3, ..., N, N is the number of oscillation modes detected.

[0022] Preferably, in step 5, the admittance sorting refers to the qq-axis admittance components. when When the value is negative, the new energy power station is the source of this broadband oscillation. This indicates that the output of the new energy power station m to this oscillation phenomenon is greater than that of the other power station k, and m should be cut off first.

[0023] The beneficial effects of this invention are that, compared with the prior art,

[0024] 1) This invention provides a method for broadband oscillation protection in power systems, which uses the positive sequence frequency ω calculated in real time by DDSRF-PLL. + Negative sequence frequency ω was excluded -The coupling effect generates second harmonics; consequently, the voltage and current on the dq axis are unaffected by harmonic interference, resulting in a more accurate admittance vector. Furthermore, compared to the common method of stability determination based on active power, using the positive sequence frequency ω... + This allows for more accurate stability assessments. In a real-world engineering project, it was found that when the system exhibits significant oscillations, the active power does not, while the positive sequence frequency reflects the dynamic characteristics of the node voltage phase angle during the oscillation process.

[0025] 2) This invention utilizes the calculated admittance of each station's output end on the dq axis to determine the magnitude of the output power for this oscillation phenomenon. Compared to the traditional analytical method in the abc three-phase coordinate system, the method of this invention avoids analyzing the two frequency components caused by the coupling of a single oscillation mode. This is because, in the dq rotating coordinate system, an oscillation mode only has a component at one frequency. Furthermore, it only utilizes the Y... qq (f n The reason for prioritizing renewable energy power plants is that the negative impedance characteristics are mainly reflected in the qq components. This approach balances the complexity and effectiveness of prioritization when switching off renewable energy units. Attached Figure Description

[0026] Figure 1 This is a flowchart of a broadband oscillation protection method for power systems according to the present invention;

[0027] Figure 2 This is a diagram of a dual-synchronous decoupling phase-locked loop structure;

[0028] Figure 3 A schematic diagram of a dual-synchronous decoupling phase-locked loop positive and negative sequence frequency decoupling module;

[0029] Figure 4 This is a schematic diagram of a new energy grid-connected system consisting of two new energy power stations in an embodiment of the present invention;

[0030] Figure 5 This is a simulation diagram of the three-phase abc voltage / current at the transmission terminals of two new energy power plants in an embodiment of the present invention;

[0031] Figure 6 This is a schematic diagram of the frequency synchronously calculated by the new energy power station 1 in an embodiment of the present invention;

[0032] Figure 7 This is a schematic diagram of the dq-axis voltage / current synchronously calculated by the new energy power station 1 in an embodiment of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0034] A method for broadband oscillation protection in power systems, such as Figure 1 As shown, it includes the following steps:

[0035] Step 1: Use a measuring device to collect the three-phase voltage and current at the output end of any new energy power station line in real time.

[0036] For example, in such Figure 4 In the shown renewable energy system, the three-phase voltage / current data of two renewable energy power stations (stations 1 and 2) on the 220kV side were measured. The three-phase voltage / current data of the two renewable energy power station lines are as follows: Figure 5 As shown, the data segment has a duration of 2 seconds and a sampling interval of 500 microseconds (sampling frequency of 2000Hz).

[0037] Step 2: Calculate the positive sequence frequency of the three-phase voltage synchronously using DDSRF-PLL to obtain the voltage and current in the dq rotating coordinate system;

[0038] The synchronous calculation of the three-phase voltage positive sequence frequency is used to transform the voltage / current from the abc stationary coordinate system to the dq rotating coordinate system; the positive sequence frequency ω of the port voltage is calculated in real time using a DDSRF-PLL (Decoupled Double Synchronous Reference-Phase Lock Loop). + and phase angle θ + And simultaneously through ω + and θ + V abc and I abc Transform to the positive sequence rotating coordinate system and

[0039] For the measured three-phase voltage, the positive and negative sequence components at the power frequency can be expressed as:

[0040]

[0041] Where +1 represents the positive-sequence power frequency component and -1 represents the negative-sequence power frequency component. Then the voltage in dq... +1 and dq -1 The coordinate axes can be represented as follows:

[0042]

[0043]

[0044] The above system of equations proves that in dq +1 The communication item in the text originates from dq -1 The coupling effect generates a harmonic component with a frequency of 2ω. To neutralize this mutual coupling, the present invention employs the method described in the appendix... Figure 2 The DDSRF-PLL structure shown is used to measure frequency.

[0045] The DDSRF-PLL architecture includes an αβ-dq conversion module:

[0046]

[0047] The DDSRF-PLL structure includes the following: Figure 3 The decoupling module shown here has n and m at +1 and -1 respectively.

[0048] The DDSRF-PLL architecture includes a low-pass filter:

[0049]

[0050] The low-pass filter parameters can be adjusted according to the bandwidth of interest in the specific system. If it is necessary to monitor high-frequency oscillations, then the low-pass filter can be omitted.

[0051] Calculated after decoupling The frequency ω' and phase angle θ' are calculated using SRF-PLL, and the positive sequence ω + =ω',θ + =θ', negative order ω - =-ω',θ - =-θ'. Similarly, it can be used to obtain...

[0052] like Figure 6 As shown, the frequency from the 5th second to the 7th second (a total of 2 seconds) is obtained using the DDSRF-PLL method described above.

[0053] Step 3: In the real-time scrolling window, determine the stability of the positive sequence frequency within the time period: if the oscillation determination condition is not met, return to step 2 to continue monitoring; if the oscillation determination condition is met, calculate the oscillation frequency and damping ratio, and continue to step 4 for analysis.

[0054] The stability determination is performed within a time window of total duration T0 = 2s, with T... s =500μs is the sampling frequency, and ω is the positive sequence frequency within the real-time time window. + Perform a stability assessment.

[0055] The stability determination method combines two factors, first determining the frequency ω within the time window. + Does the difference between local extrema exceed the warning value ±Δf? lim =1 (Hz), and the number of occurrences exceeds N. p =2; Since the condition is met, namely, the positive sequence frequency local extrema appear consecutively for more than 1.7Hz, the ERA (Eigenvalue Realization Algorithm) algorithm is activated to calculate the positive sequence frequency ω. + The oscillation mode frequencies (f1, f2, ..., f) in N ) and damping ratio (D1,D2,…,D N N represents the number of monitored oscillation modes. Stability is determined based on the oscillation mode damping ratio.

[0056] ERA can calculate a linear time-invariant (LTI) discrete system based on the frequency measured within this 2-second time window:

[0057] x k+1 =Ax k +Bu k y k =Cx k +Du k

[0058] Since ERA assumes the input is an impulse signal, it cannot evaluate an I / O model; instead, it estimates a frequency-domain state equation defined by matrices A, B, C, and D based on the measured output signal. Its order needs to be defined beforehand. The measured signal in the frequency domain can be represented as:

[0059]

[0060] Where A′ is the system matrix of the continuous dynamic equation, V is its right eigenvector, and Ω is its diagonal matrix.

[0061] For the k-th measurement signal, it can be represented in the frequency domain as:

[0062]

[0063] A key step in ERA is constructing two Hankel matrices:

[0064]

[0065]

[0066] Therefore, the Hankel matrix can be decomposed into:

[0067]

[0068] in It is the observability matrix. It is the controllability matrix. ERA is obtained using Singular Value Decomposition (SVD) and order reduction. and Therefore, matrix A can be found through H1 and H2.

[0069] Furthermore, the eigenvalue decomposition of matrix A yields the oscillation mode frequencies (f1, f2, ..., f...). N ) and damping ratio (D1, D2, ..., D N N represents the number of monitored oscillation modes. Stability is determined based on the oscillation mode damping ratio.

[0070] In any f n Under the corresponding mode n, if the damping ratio satisfies 0% <D n <D lim If the system is considered to be at critical stability with a risk of oscillation, an early warning can be issued, and the power generation station can be temporarily not shut down while monitoring continues; if the oscillation mode damping ratio satisfies D... n If the percentage is ≤0%, it indicates that the system has experienced wideband oscillations, triggering an early warning and preparing for the next step of disconnecting new energy power plants.

[0071] In this case, a primary oscillation mode was detected with a frequency of f1 = 6.28 Hz and a corresponding damping ratio of D1 = -0.39%. This indicates that the system is experiencing broadband oscillation, and an early warning should be issued. The following steps should then be taken to determine the renewable energy power plants that should be prioritized for shutdown.

[0072] Step 4: Calculate the admittance of the corresponding oscillation mode on the dq axis using the voltage and current on the dq axis;

[0073] At this point, the judgment is that there is an oscillation problem, and step 2 has already been completed to obtain voltage / current data with a time duration of 2 seconds and a sampling interval of 500 microseconds. Then, the vector corresponding to the oscillation mode f1 = 6.28Hz, i.e., V, is calculated using FFT. d (f1) and V q (f1), similarly calculate the current vector I. d (f1) and I q (f1), and calculate the admittance matrix on the dq axis according to the following formula:

[0074]

[0075] Where m = 1, 2, representing the two new energy power stations required in the system of this embodiment.

[0076] Step 5: Determine whether each new energy power station is an oscillation source by the admittance of its transmission end at the oscillation frequency, sort the admittances, identify the most serious new energy power station, and cut it off.

[0077] When broadband oscillations occur in new energy systems, they are usually very small relative to the power frequency. Therefore, linearizable ERA evaluation algorithms and dq-axis admittance algorithms can be used. Moreover, broadband oscillation problems mainly manifest in the positive-sequence component because common broadband oscillation problems do not exhibit three-phase asymmetry.

[0078] Furthermore, the reason for transforming the voltage and current to the dq rotating coordinate system for analysis is to avoid analyzing the two sub-supersynchronous components generated by the coupling of a single oscillation mode in the abc coordinate system. In the dq rotating coordinate system, only the component at one frequency needs to be considered.

[0079] Currently, extensive research has shown that the negative impedance characteristic of new energy systems is due to the negative impedance characteristics of the qq component. The dd impedance still exhibits positive impedance characteristics similar to those of a current source. The dq and qd impedances are relatively very small, while the qq component is influenced by parameters such as the PLL and current control loop, resulting in a relatively large negative impedance, especially under certain operating conditions.

[0080] Therefore, this invention only considers the sorting of the qq admittance vectors, making it more efficient in selecting broadband oscillation sources. The method involves determining the qq-axis admittance components of any given station. when When the value is negative, then the new energy power station m is the source of this broadband oscillation. This indicates that the output of the new energy power station m to this oscillation phenomenon is greater than that of the other power station k, and m should be cut off first.

[0081] In this case, sorting the qq component admittances yields the following results: Right now This indicates that power station 1 has a greater output at 6.28Hz oscillation compared to power station 2. Therefore, it is recommended to shut down the new energy power station 1 first.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for broadband oscillation protection in new energy power systems, characterized in that, Includes the following steps: Step 1: Use a measuring device to collect the three-phase voltage and current at the output end of any new energy power station line in real time. Step 2: Calculate the positive sequence frequency of the three-phase voltage synchronously using DDSRF-PLL to obtain the voltage and current in the dq rotating coordinate system; Step 3: In the real-time scrolling window, determine the stability of the positive sequence frequency within the time period: If the oscillation determination condition is not met, return to Step 2 to continue monitoring; if the oscillation determination condition is met, calculate the oscillation frequency and damping ratio, and continue to Step 4 for analysis. Step 4: Calculate the admittance of the oscillation mode on the dq axis using the voltage and current on the dq axis; Step 5: Determine whether each new energy power station is an oscillation source by the admittance of its transmission end at the oscillation frequency, sort the admittances, identify the most serious new energy power station, and cut it off.

2. The method for broadband oscillation protection of new energy power systems according to claim 1, characterized in that: In step 3, the stability determination is based on the total duration being Within the time window, The sampling frequency is the positive sequence frequency within the real-time time window. Perform a stability assessment.

3. The method for broadband oscillation protection of new energy power systems according to claim 2, characterized in that: The stability determination method combines two factors: first, it determines the frequency within the time window. Does the difference between local extrema exceed the warning value? (Hz), and the number of occurrences exceeds If the conditions are met, the feature system implementation algorithm is activated to calculate the positive sequence frequency. Oscillation mode frequency in Damping ratio , , The number of oscillation modes detected; Secondly, stability is determined based on the damping ratio of the oscillation modes; In any Under the corresponding mode n, if the damping ratio satisfies If the system is considered to be at critical stability with a risk of oscillation, an early warning is issued, and the renewable energy power station can be temporarily not shut down while monitoring continues; if the oscillation mode damping ratio satisfies This indicates that the system has experienced wideband oscillations, triggering an early warning and preparing for the next step of disconnecting new energy power plants.

4. The method for broadband oscillation protection of new energy power systems according to claim 1, characterized in that: In step 4, the admittance corresponding to the oscillation mode on the dq axis is calculated using the voltage and current on the dq axis, and the corresponding oscillation mode is calculated on the dq axis. The voltage / current vector under the current, i.e. , , and And calculate the admittance matrix on the dq axis according to the following formula: in, , The total number of new energy power stations measured; , The number of oscillation modes detected.

5. A method for broadband oscillation protection of new energy power systems according to claim 1, characterized in that: In step 5, the admittance sorting is the q-axis self-admittance component. ,when When the value is negative, the new energy power station is the source of this broadband oscillation. This indicates a new energy power station. The power output relative to another station for this oscillation phenomenon Larger ones should be removed first. .

Images