A broadband oscillation monitoring method and device

Abstract

This invention discloses a broadband oscillation monitoring method and apparatus. The method includes: segmenting and preprocessing real-time acquired electrical measurement data in the range of 0–2500 Hz; extracting features from signals within a sliding window based on the results of the segmented preprocessing; and predicting the broadband oscillation state using a fully connected network (FC) based on the extracted feature values. The solution of this invention performs segmented preprocessing on real-time acquired electrical measurement data, then uses sliding window feature extraction to complete FC-based state prediction, thereby discriminating the monitored electrical measurement data and achieving rapid identification of broadband oscillations.

Description

Technical Field

[0001] This invention relates to the field of power system automation technology, and specifically to a broadband oscillation monitoring method and device. Background Technology

[0002] With the large-scale grid connection of new energy sources, the modern power system is characterized by "high energy consumption and high power density." The interaction between power electronic equipment and the power grid in such systems can cause broadband oscillations ranging from a few hertz to several kilohertz. These broadband oscillations can damage power equipment and lead to the shutdown of new energy generating units. The impact of these oscillations can extend to multiple units and electrical equipment across multiple regions. On the one hand, they may affect the safety and stability of the power electronic equipment itself, thereby affecting the reliable power generation and transmission of new energy sources. On the other hand, the oscillations may also interact with other equipment in the power grid, triggering global safety and stability issues.

[0003] Wideband oscillations have the following characteristics: 1) Their mechanism involves the dynamic interaction between multiple converters, new energy units, and AC / DC power grids, which is fundamentally different from traditional motor oscillations; 2) They have a wide oscillation frequency range, from a few hertz to several kilohertz, encompassing both mechanical torsional vibration and electrical oscillation frequencies, posing a risk of triggering resonance; 3) The frequency, damping, or stability of the oscillations is affected by numerous parameters of the converter and power grid, as well as external conditions such as wind and solar power, exhibiting complex influencing factors and a wide range of time-varying characteristics; 4) Converters have complex nonlinear characteristics and limited overload capacity, and control signals are easily limited, causing oscillations to often begin with small-signal negative damping divergence and end with continuous nonlinear oscillations. Therefore, it is necessary to identify and address wideband oscillations.

[0004] Currently, WAMS (Wide Area Measurement System) utilizes fundamental phasor measurement data from PMU (Phasor Measurement Unit) to monitor subsynchronous oscillations below 50Hz, but it lacks real-time monitoring capabilities. Even with a maximum transmission rate of 100 frames per second, WAMS cannot achieve real-time monitoring of supersynchronous oscillations above 50Hz or even broadband oscillations. Therefore, WAMS cannot address the challenges of broadband oscillations. Furthermore, since many broadband oscillations exhibit non-power frequency signals or harmonic signals, with the vast majority falling into the interharmonic range, and PMU devices are only used for measuring the fundamental phasor of the power grid, they cannot meet the real-time monitoring requirements of interharmonic signals. Therefore, it is necessary to propose new technologies to achieve real-time monitoring of broadband oscillations. Summary of the Invention

[0005] The purpose of this invention is to provide a broadband oscillation monitoring method and apparatus for real-time monitoring of broadband oscillations.

[0006] To achieve the above objectives, the first aspect of the present invention discloses a broadband oscillation monitoring method, comprising:

[0007] The electrical measurement data acquired in real time within the range of 0–2500 Hz are preprocessed in segments.

[0008] Feature extraction is performed on the signal within the sliding window based on the results of segmented preprocessing.

[0009] Wideband oscillation state prediction is performed using a fully connected network (FC) based on the extracted feature values.

[0010] Optionally, the segmented preprocessing of the real-time acquired electrical measurement data in the 0–2500Hz range includes:

[0011] For the 0–100 Hz frequency band, the electrical measurement data are subjected to low-pass filtering and windowing to obtain the low-frequency component;

[0012] For the 100–2500 Hz frequency band, the fundamental frequency is tracked based on the electrical measurement data, and the high-frequency component is obtained by resampling.

[0013] Optionally, the feature extraction of the signal within the sliding window based on the results of segmented preprocessing includes:

[0014] For the high-frequency components of the segmented preprocessing, multiple types of features are extracted from the signal within the sliding window by using Fast Fourier Transform (FFT) and spectral grouping.

[0015] The multiple types of features include: harmonic features and interharmonic features.

[0016] Optionally, before performing multi-type feature extraction on the signal within the sliding window, the method further includes:

[0017] The spectral lines of the high-frequency components are corrected;

[0018] Accordingly, multi-type feature extraction is performed on the calibrated signal within the sliding window.

[0019] Optionally, the broadband oscillation state prediction based on the extracted feature values ​​using a fully connected network includes:

[0020] For each type of feature, feature values ​​from multiple consecutive windows are taken to form a continuous window feature combination.

[0021] Autoencoder (AE) is used to mine continuous window features of multiple types of features.

[0022] The difference between the results of AE mining and the combination of continuous window features determines whether a broadband oscillation event has occurred. By splicing together multiple types of continuous window feature combinations, FC is used as a discriminator. The final output of FC is the probability that the signal in the current sliding window will oscillate broadbandly.

[0023] Optionally, the method further includes: when a broadband oscillation is detected in more than or equal to N sliding windows out of M consecutive sliding windows, it is determined that a broadband oscillation has occurred; otherwise, no broadband oscillation has occurred, where M is greater than N.

[0024] Optionally, the method further includes: performing sub / supersynchronous monitoring via windowed FFT, spectral correction, and dominant components.

[0025] A second aspect of the present invention discloses a broadband oscillation monitoring device, comprising:

[0026] The segmented preprocessing module is used to perform segmented preprocessing on the electrical measurement data in the real-time range of 0 to 2500 Hz.

[0027] The window feature extraction module is used to extract features from the signal within the sliding window based on the results of segmented preprocessing.

[0028] A wideband oscillation monitoring module is used to predict wideband oscillation states using a fully connected network (FC) based on extracted feature values.

[0029] A third aspect of this invention discloses an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of a broadband oscillation monitoring method according to any one of the first aspects of this disclosure.

[0030] A fourth aspect of this invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a broadband oscillation monitoring method according to any one of the first aspects of this disclosure.

[0031] The solution provided by this invention preprocesses the electrical measurement data in the range of 0-2500Hz collected in real time by segmentation, and then performs state prediction based on FC by using feature extraction of sliding window, so as to identify the monitored electrical measurement data and realize the function of rapid identification of broadband oscillation monitoring. Attached Figure Description

[0032] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0033] Figure 1 This is a flowchart illustrating a broadband oscillation monitoring method according to an embodiment of the present invention;

[0034] Figure 2 This is a schematic diagram illustrating the measurement of harmonics and interharmonics using a grouping model according to an embodiment of the present invention.

[0035] Figure 3 This is a schematic diagram of a subsynchronous / supersynchronous monitoring method according to an embodiment of the present invention;

[0036] Figure 4 This is a schematic diagram of a broadband oscillation monitoring neural network according to an embodiment of the present invention;

[0037] Figure 5 This is a schematic diagram of a broadband oscillation monitoring device according to an embodiment of the present invention;

[0038] Figure 6 This is a schematic diagram of a broadband oscillation monitoring system according to an embodiment of the present invention;

[0039] Figure 7 This is a schematic diagram of another broadband oscillation monitoring system according to an embodiment of the present invention;

[0040] Figure 8 This is a structural diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0041] The technical solution of the present invention will be clearly and completely described below with reference to specific embodiments. However, those skilled in the art should understand that the embodiments described below are only for illustrating the present invention and should not be regarded as limiting the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Example 1:

[0043] The first aspect of this invention discloses a broadband oscillation monitoring method. Figure 1 This is a flowchart illustrating a broadband oscillation monitoring method according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0044] S101. Perform segmented preprocessing on the electrical measurement data in the real-time range of 0-2500Hz;

[0045] S102. Based on the results of segmented preprocessing, feature extraction is performed on the signal within the sliding window;

[0046] S103. Based on the extracted feature values, use a fully connected network (FC) to predict broadband oscillation states.

[0047] In step S101, electrical measurement data within the 0–2500Hz range is first acquired in real time. This electrical measurement data includes, but is not limited to, voltage, current, and time-stamped signals. Subsequently, the acquired electrical measurement data within the 0–2500Hz range is preprocessed in segments, and a synchronization phasor is calculated based on a preset high sampling rate. This high sampling rate can be several times higher than that of a conventional PMU device, thereby achieving higher phasor measurement accuracy, for example, three times higher.

[0048] Specifically, in some embodiments of the present invention, for sub / supersynchronous signals in the 0–100 Hz frequency band, the electrical measurement data can be low-pass filtered and windowed to obtain low-frequency components; for interharmonic signals in the 100–2500 Hz frequency band, the fundamental frequency is tracked based on the electrical measurement data, and high-frequency components are obtained through resampling. For the segmented preprocessed low-frequency components, sub / supersynchronous monitoring can be performed using techniques commonly used by those skilled in the art, which will not be detailed here.

[0049] In some embodiments of the present invention, for the 0–100 Hz frequency band, the subsynchronous / supersynchronous components can be obtained by decimating the sampled data after low-pass filtering. In practical applications, the frequency resolution of the subsynchronous / supersynchronous components can reach 0.5 Hz. In the low-frequency band, frequency aliasing and spectral leakage are eliminated based on FIR (Finite Impulse Response) filtering and windowing algorithms, achieving high-frequency resolution measurement. In some embodiments, the Hanning window function, suitable for non-periodic continuous signals, can be used. It is understood that the specific schemes of the FIR filtering decimation scheme and windowing algorithm can adopt schemes commonly used by those skilled in the art, and will not be detailed here.

[0050] In some embodiments of the present invention, for the 100–2500 Hz frequency band, resampled data obtained by interpolating the fundamental frequency can be tracked to obtain interharmonic components. In practical applications, the frequency resolution of interharmonic components can reach 5 Hz (1 / 10 of the fundamental frequency). In tracking the fundamental frequency in the high-frequency band, a short data window is generated by resampling to ensure the algorithm's response speed to dynamic shifts in the fundamental frequency, thereby suppressing the impact of harmonic spectral leakage on interharmonic measurements. It is understood that the generation of the short data window can adopt methods commonly used by those skilled in the art, which will not be detailed here.

[0051] Different frequency bands represent different information. By processing real-time electrical measurement data in different frequency bands, differentiated processing makes different frequency bands pay more attention to the occurrence of oscillation events.

[0052] In step S102, feature extraction is performed on the signal within the sliding window based on the results of segmented preprocessing.

[0053] In some embodiments of the present invention, for the high-frequency components of segmented preprocessing, multi-type feature extraction is performed on the signal within the sliding window according to FFT (Fast Fourier Transform) and spectral grouping; wherein, the multi-type features include: harmonic features and interharmonic features.

[0054] Figure 2 This is a schematic diagram illustrating the measurement of harmonics and interharmonics using a grouping model according to an embodiment of the present invention, as shown below. Figure 2 As shown, after data segmentation preprocessing via a preprocessing data window, harmonic and interharmonic features can be extracted based on high-frequency components using FFT and spectral grouping, respectively. Specifically, harmonic and interharmonic features can be defined as the root-mean-square (RMS) values ​​of harmonics and interharmonics. In practical applications, the RMS values ​​of harmonic subgroups and interharmonic central subgroups can be calculated separately using an FFT spectral grouping strategy. Furthermore, in addition to harmonic and interharmonic RMS values, other types of features, such as energy spectra, can be added for feature extraction.

[0055] Furthermore, in some embodiments of the present invention, the spectral lines of the high-frequency components are corrected before multi-type feature extraction of the signal within the sliding window; correspondingly, multi-type feature extraction is performed on the corrected signal within the sliding window. Full-band spectral line correction from 100 to 2500 Hz facilitates rapid analysis and location of steady-state harmonic and interharmonic current distributions and directions. Thus, through the root mean square values ​​of harmonics, root mean square values ​​of interharmonics, and frequency estimation, harmonic and interharmonic energy can be rapidly assessed and located in real time. Simultaneous monitoring and alarming of harmonic and interharmonic events can be achieved, while minimizing the amount of data transmitted.

[0056] In practical applications, a grouping model is used to measure harmonics and interharmonics, with a sliding window time window of 10 cycles (≈200ms). This step uses a 200ms data window, which is compatible with current measurement and control device specifications regarding data windows and can replace conventional SCADA (Supervisory Control And Data Acquisition) measurements.

[0057] Regarding sliding window signal feature extraction, in some embodiments of the present invention, the window size and sliding distance of the sliding window can be preset. For example, the window size can be set to 10 cycles, and the sliding distance of each cycle can be set to 1 cycle. Then, multiple types of features can be extracted from the signal within the sliding window. These extracted features include harmonic features and interharmonic features. In practical applications, the root mean square value of the h-th harmonic, the root mean square value of the h-th interharmonic, and the signal energy within the window can be extracted. By utilizing the root mean square values ​​of the h-th and h-th interharmonics, along with the energy spectrum and other features, a comprehensive evaluation of broadband oscillations is achieved, reducing reliance on a single feature and minimizing the occurrence of false detection events related to broadband oscillations caused by a single feature.

[0058] Furthermore, in some embodiments of the present invention, the frequency, amplitude, and phase of spectral lines in the 100–2500 Hz frequency band of a signal can be used to analyze and locate the distribution and direction of steady-state harmonic / interharmonic currents. In practical applications, spectrum correction requires a data window of seconds in length and is affected by averaging effects, making it suitable for measuring steady-state signals. Figure 3 This is a schematic diagram of a sub / supersynchronous monitoring method according to an embodiment of the present invention, as shown below. Figure 3 As shown, subsynchronous / supersynchronous monitoring can be performed through windowed FFT, spectral correction, and dominant components.

[0059] Specifically, we can first solve for the narrow-band equivalent values ​​of the subsynchronous and supersynchronous currents and voltages under non-sinusoidal and unbalanced conditions and the distortion power formed by their interaction, and monitor the energy of the subsynchronous / supersynchronous components from an overall perspective.

[0060] Among them, the equivalent value I of the phase current sub / supersynchronous component peH Defined as follows:

[0061]

[0062] Oscillation power P eH Defined as follows:

[0063] P eH =3U e1 I eH

[0064] In the formula: I pThe effective value of phase current, I p1 This represents the effective value of the fundamental current. Monitoring of narrowband energy in the 2.5–45 Hz and 55–95 Hz ranges, U e1 I eH The fundamental voltage and secondary / supersynchronous current components after three-phase synthesis are given, with zero-sequence oscillation power components taken into account.

[0065] Furthermore, based on the windowed FFT spectrum, the frequency, amplitude, and phase correction of the dominant components of the steady-state signal are achieved. Sampling synchronized with the clock determines the direction of the sub- / super-synchronous current and oscillation power components. The aim is to solve for the dominant components of current, voltage, and instantaneous power in the sub- / super-synchronous frequency domain under steady-state conditions, including frequency, amplitude, and phase, providing auxiliary information for oscillation source localization.

[0066] The frequency correction formula is shown below:

[0067]

[0068] The amplitude correction formula is shown below:

[0069]

[0070] In the formula: f s / N represents the frequency, G represents the spectral line energy, k represents the spectral line number of the peak, m and n represent the number of spectral lines deviating from the peak at the edge, and K represents the spectral line number. t A is the energy recovery coefficient of the window function, and A is the amplitude.

[0071] Furthermore, by applying windowed FFT correction, and by combining point and surface monitoring of the subsynchronous and supersynchronous components of current, voltage, and instantaneous power, accurate measurement of the frequency, amplitude, and phase of the dominant components of voltage, current, and instantaneous power is achieved.

[0072] In step S103, broadband oscillation state prediction is performed using a fully connected network (FC) based on the extracted feature values.

[0073] In some embodiments of the present invention, for each type of feature, feature values ​​from multiple consecutive windows are taken to form a continuous window feature combination; an autoencoder (AE) is used to mine the continuous window feature combination of multiple types of features; the difference between the AE mining result and the continuous window feature combination is used to determine whether a broadband oscillation event has occurred. After concatenating the continuous window feature combinations of multiple types, a current encoder (FC) is used as a discriminator, and the final output of the FC is the probability of broadband oscillation of the signal within the current sliding window. In practical applications, the signal used is the signal that has been corrected in the previous step. Figure 4 This is a schematic diagram of a wideband oscillation monitoring neural network according to an embodiment of the present invention, as shown below. Figure 4As shown, wideband oscillations can be monitored using sliding windows, autoencoder networks, and fully connected networks.

[0074] In some embodiments of the present invention, after feature extraction of the signal within the sliding window, feature values ​​from multiple consecutive windows can be taken for each type of feature to form a continuous window feature combination. Specifically, the number of windows within the continuous window used for feature combination can be preset. Then, for any type of feature, feature values ​​of the signals within multiple consecutive windows are taken respectively for that type of feature to form a one-dimensional feature vector, i.e., the continuous window feature combination of that type of feature. For example, signals from 20 consecutive windows can be combined. Harmonic features, interharmonic features, and energy features corresponding to the continuous window signals are taken, each forming a one-dimensional feature vector. In practical applications, considering that the signal is a continuous signal and the sliding window is continuous, the signal features maintain correlation and can reflect the trend of the signal. Therefore, a recurrent neural network (RNN) can be used to extract relevant features.

[0075] Furthermore, considering the correlation between different features, such as the correlation between the simple harmonic characteristics and harmonic characteristics of the same signal, the correlation between multiple features can be used to achieve comprehensive evaluation of broadband oscillations. In some embodiments of the present invention, in practical applications, the function concat can be used to concatenate multidimensional vectors.

[0076] Considering the use of Auto Encoder (AE) networks, hidden information for each type of feature can be further mined. Therefore, in some embodiments of this invention, AE mining can be performed on continuous window feature combinations of multiple feature types using an autoencoder (AE). Since the basic idea of ​​an AE network is to train the hidden layer's ability to extract features using the same input and output, the hidden layer contains all the information of the original input. In practical applications, information features under non-oscillatory conditions can be used as the input and output of the AE network, enabling it to learn the relevant feature extraction methods under non-oscillatory conditions. For rare wide-screen oscillation anomalies, the AE network's ability to express oscillations is weak. In this case, with an input containing wide-screen oscillations, the output will reduce the information related to wide-screen oscillations after passing through the AE network. Thus, the difference between the input and output can be used to determine whether a wide-frequency oscillation event has occurred. In this way, by using the AE autoencoder network to learn the expression of relevant features under non-oscillatory conditions, and then using the difference in features to further determine the probability of wide-frequency oscillations occurring when they occur, it is possible to effectively monitor wide-frequency oscillation signal waveforms that have not yet appeared. AE improves the network's generalization ability for wide-frequency oscillation monitoring.

[0077] Considering that the impact of oscillations on signal features is greatly reduced after multiple types of features pass through an autoencoder (AE), in some embodiments of this invention, the difference between the AE mining result and the combination of continuous window features is used to determine whether a broadband oscillation event has occurred. This is achieved by concatenating multiple types of continuous window feature combinations and using a discriminator (FC). The final output of the FC is the probability that the signal within the current sliding window has experienced a broadband oscillation. For example, if the probability p > 0.5, a broadband oscillation is considered to have occurred; otherwise, it has not.

[0078] Furthermore, considering that the sliding window is continuous, the window size is larger than the sliding distance, and the period of the oscillation is relatively short, the same broadband oscillation is usually detected by multiple consecutive sliding windows.

[0079] Preferably, in some embodiments of the present invention, in order to improve the correct detection rate of broadband oscillation detection and reduce the false detection rate, a broadband oscillation alarm strategy can be set: when broadband oscillation is detected in more than or equal to N sliding windows out of M consecutive sliding windows, it is determined that broadband oscillation has occurred; otherwise, broadband oscillation has not occurred, where M is greater than N. For example, if each sliding distance is only 1 cycle and the window size is 10 cycles, the following broadband oscillation alarm strategy can be set: when broadband oscillation is detected in more than or equal to 3 windows out of 5 consecutive sliding windows, it is considered that broadband oscillation has occurred; otherwise, broadband oscillation has not occurred.

[0080] The solution provided by this invention preprocesses the electrical measurement data in the range of 0-2500Hz collected in real time by segmentation, and then performs state prediction based on FC by using feature extraction of sliding window, so as to identify the monitored electrical measurement data and realize the function of rapid identification of broadband oscillation monitoring.

[0081] Example 2:

[0082] This invention discloses a broadband oscillation monitoring device. Figure 5 This is a schematic diagram of a broadband oscillation monitoring device according to an embodiment of the present invention; as shown. Figure 5 As shown, the broadband oscillation monitoring device 500 includes: a segmented preprocessing module 501, a window feature extraction module 502, and a broadband oscillation monitoring module 503.

[0083] The segmented preprocessing module 501 is used to perform segmented preprocessing on the electrical measurement data in the 0-2500Hz range acquired in real time.

[0084] The window feature extraction module 502 is used to extract features from the signal within the sliding window based on the results of segmented preprocessing.

[0085] The wideband oscillation monitoring module 503 is used to predict the state of wideband oscillations using a fully connected network (FC) based on the extracted feature values.

[0086] In some embodiments of the present invention, for the 0-100Hz frequency band, the segmented preprocessing module 501 performs low-pass filtering and windowing processing on the electrical measurement data to obtain low-frequency components; for the 100-2500Hz frequency band, the segmented preprocessing module 501 tracks the fundamental frequency based on the electrical measurement data and obtains high-frequency components through resampling.

[0087] In some embodiments of the present invention, for the high-frequency components of segmented preprocessing, the window feature extraction module 502 performs multi-type feature extraction on the signal within the sliding window according to the Fast Fourier Transform (FFT) and spectral grouping; wherein, the multi-type features include: harmonic features and interharmonic features.

[0088] In some embodiments of the present invention, the window feature extraction module 502 corrects the spectral lines of the high-frequency component before performing multi-type feature extraction on the signal within the sliding window; correspondingly, multi-type feature extraction is performed on the corrected signal within the sliding window.

[0089] In some embodiments of the present invention, for each type of feature, the broadband oscillation monitoring module 503 takes feature values ​​of multiple consecutive windows to form a continuous window feature combination; the broadband oscillation monitoring module 503 performs AE mining on the continuous window feature combination of multiple types of features through an autoencoder AE, and determines whether a broadband oscillation event has occurred based on the difference between the AE mining result and the continuous window feature combination; after splicing the continuous window feature combinations of multiple types, the FC is used as a discriminator, and the final output of the FC is the probability of broadband oscillation of the signal in the current sliding window.

[0090] Furthermore, in some embodiments of the present invention, when more than or equal to N sliding windows out of M consecutive sliding windows detect the occurrence of broadband oscillation, the broadband oscillation monitoring module 503 determines that broadband oscillation has occurred; otherwise, no broadband oscillation has occurred, wherein M is greater than N.

[0091] Furthermore, in some embodiments of the present invention, the broadband oscillation monitoring module 503 performs sub / supersynchronous monitoring through windowed FFT, spectrum correction, and dominant component.

[0092] It is understood that the specific functional implementation of each module in the broadband oscillation monitoring device provided in Embodiment 2 of this disclosure can refer to the specific implementation scheme of each step of the broadband oscillation monitoring method provided in Embodiment 1, and will not be repeated here.

[0093] The broadband oscillation monitoring device provided by the present invention preprocesses the electrical measurement data in the range of 0 to 2500 Hz in real time by segmentation, and then performs state prediction based on FC by using feature extraction of sliding window, so as to identify the monitored electrical measurement data and realize the function of rapid identification of broadband oscillation monitoring.

[0094] Example 3:

[0095] This invention discloses a broadband oscillation monitoring system. Figure 6 This is a schematic diagram of a broadband oscillation monitoring system according to an embodiment of the present invention; as shown. Figure 6 As shown, the system includes: a broadband master station, a broadband measurement device, a broadband communication unit, and a time synchronization device.

[0096] Figure 6 In electrical circuits, a PT (Potential Transformer) is an instrument used to transform voltage. The primary purpose of a voltage transformer is to power measuring instruments and relay protection devices, to measure line voltage, power, and energy, or to protect valuable equipment, motors, and transformers in the event of a line fault. A CT (Current Transformer) is an instrument that measures current by converting a large primary current into a small secondary current based on the principle of electromagnetic induction. A time synchronization device is used to ensure time synchronization of broadband measuring devices and broadband communication units.

[0097] The broadband master station collects alarm and diagnostic analysis files in real time and monitors broadband electrical signals. It allows viewing of waveform recordings, spectrum analysis diagrams, and operation reports for each bay, and monitors and statistically analyzes the type, frequency, and timing of fault alarms in each bay of the new energy plant. This provides a reference for future fault early warning. For bays with frequent alarms, daily, weekly, and monthly reports are generated, statistically analyzing typical fault types and the frequency of fault occurrences.

[0098] The broadband communication unit stores and records broadband measurement data and long waveform data; performs station-domain analysis of operational status statistics and offline / online data; and provides communication functions, transmitting real-time data, alarm diagnostic results, and log files.

[0099] A wideband measurement device that acquires high-frequency signals in real time; performs real-time measurements of fundamental, harmonics, and interharmonics; and completes alarms and waveform recording for sub / supersynchronous oscillations, interharmonics, and harmonic over-limits.

[0100] It is understood that the broadband measurement device is the broadband oscillation monitoring device provided in Example 2.

[0101] In some embodiments of the present invention, the broadband oscillation monitoring system may deploy one or more broadband measurement devices. Figure 7 This is a schematic diagram of another broadband oscillation monitoring system according to an embodiment of the present invention; as shown. Figure 7 As shown, multiple broadband measurement devices are deployed through a switch in the broadband oscillation monitoring system.

[0102] This invention provides a broadband oscillation monitoring system. By selecting some new energy power plants connected to weak AC power grids and applying broadband oscillation monitoring devices, the monitoring, analysis and handling capabilities of power plants for broadband oscillations can be effectively improved.

[0103] Example 4:

[0104] This invention discloses an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of a broadband oscillation monitoring method according to any one of the embodiments disclosed in 1 of this invention.

[0105] Figure 8 This is a structural diagram of an electronic device according to an embodiment of the present invention, such as... Figure 8 As shown, the electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, Near Field Communication (NFC), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0106] Those skilled in the art will understand that Figure 8 The structure shown is merely a structural diagram of the part related to the technical solution of this disclosure and does not constitute a limitation on the electronic device to which the solution of this application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0107] Example 5:

[0108] This invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a broadband oscillation monitoring method according to Embodiment 1 of this invention.

[0109] Please note that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. The above embodiments only illustrate several implementation methods of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

[0110] The embodiments of the subject matter and functional operation described in this specification can be implemented in the following ways: digital electronic circuits, tangibly embodied computer software or firmware, computer hardware including the structures disclosed in this specification and their structural equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible, non-transitory program carrier for execution by a data processing apparatus or for controlling the operation of a data processing apparatus. Alternatively or additionally, the program instructions may be encoded on artificially generated propagation signals, such as machine-generated electrical, optical, or electromagnetic signals, which are generated to encode information and transmit it to a suitable receiving device for execution by the data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or combinations thereof.

[0111] The processing and logic flow described in this specification can be executed by one or more programmable computers that execute one or more computer programs to perform corresponding functions by operating on input data and generating output. The processing and logic flow can also be executed by dedicated logic circuitry—such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits), and the device can also be implemented as dedicated logic circuitry.

[0112] Suitable computers for executing computer programs include, for example, general-purpose and / or special-purpose microprocessors, or any other type of central processing unit. Typically, the central processing unit receives instructions and data from read-only memory and / or random access memory. The basic components of a computer include a central processing unit for implementing or executing instructions and one or more memory devices for storing instructions and data. Typically, a computer will also include one or more mass storage devices for storing data, such as disks, magneto-optical disks, or optical disks, or the computer will be operatively coupled to such mass storage devices to receive data from or transfer data to them, or both. However, a computer is not required to have such devices. Furthermore, a computer can be embedded in another device, such as a mobile phone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a global positioning system (GPS) receiver, or a portable storage device such as a universal serial bus (USB) flash drive, to name a few.

[0113] Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and memory devices, such as semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROM and DVD-ROM disks. Processors and memory may be supplemented by or incorporated into dedicated logic circuitry.

[0114] While this specification contains numerous specific implementation details, these should not be construed as limiting the scope of any invention or the scope of the claims, but rather are primarily intended to describe features of specific embodiments of a particular invention. Certain features described in the various embodiments herein may also be implemented in combination in a single embodiment. Conversely, various features described in a single embodiment may also be implemented separately in various embodiments or in any suitable sub-combination. Furthermore, while features may function in certain combinations as described above and even initially claimed in this way, one or more features from a claimed combination may be removed from that combination in some cases, and a claimed combination may refer to a sub-combination or a variation thereof.

[0115] Similarly, although the operations are depicted in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order shown or sequentially, or requiring all illustrated operations to be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system modules and components in the above embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0116] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims may be performed in a different order and still achieve the desired result. Furthermore, the processes depicted in the drawings are not necessarily shown in a specific order or sequence to achieve the desired result. In some implementations, multitasking and parallel processing may be advantageous.

[0117] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A broadband oscillation monitoring method, characterized in that, include: The electrical measurement data acquired in real time within the range of 0~2500Hz are preprocessed in segments; Feature extraction is performed on the signal within the sliding window based on the results of segmented preprocessing. Wideband oscillation state prediction is performed using a fully connected network (FC) based on the extracted feature values. The segmented preprocessing of the real-time acquired electrical measurement data in the 0~2500Hz range includes: For the 0~100Hz frequency band, the electrical measurement data is subjected to low-pass filtering and windowing to obtain the low-frequency component; For the 100~2500Hz frequency band, the fundamental frequency is tracked based on the electrical measurement data, and the high-frequency component is obtained by resampling; The feature extraction of the signal within the sliding window based on the results of segmented preprocessing includes: For the high-frequency components of the segmented preprocessing, multiple types of features are extracted from the signal within the sliding window by using Fast Fourier Transform (FFT) and spectral grouping. The multiple types of features include: harmonic features and interharmonic features. Before performing multi-type feature extraction on the signal within the sliding window, the method further includes: The spectral lines of the high-frequency components are corrected; Accordingly, multi-type feature extraction is performed on the calibrated signal within the sliding window; Furthermore, the broadband oscillation state prediction based on the extracted feature values ​​using a fully connected network (FC) includes: For each type of feature, feature values ​​from multiple consecutive windows are taken to form a continuous window feature combination. Autoencoder (AE) is used to mine continuous window features of multiple types of features. The difference between the results of AE mining and the combination of continuous window features determines whether a broadband oscillation event has occurred. By splicing together multiple types of continuous window feature combinations, FC is used as a discriminator. The final output of FC is the probability that the signal in the current sliding window will oscillate broadbandly.

2. The broadband oscillation monitoring method according to claim 1, characterized in that, Also includes: If a wideband oscillation is detected in M ​​consecutive sliding windows, and N or more sliding windows detect the occurrence of wideband oscillation, then a wideband oscillation is determined to have occurred; otherwise, no wideband oscillation has occurred. Here, M is greater than N.

3. The broadband oscillation monitoring method according to claim 1, characterized in that, Also includes: Subsynchronous / supersynchronous monitoring is performed using windowed FFT, spectral correction, and dominant components.

4. A broadband oscillation monitoring device, utilizing the broadband oscillation monitoring method of claim 1, characterized in that, include: The segmented preprocessing module is used to perform segmented preprocessing on the electrical measurement data in the real-time range of 0~2500Hz. The window feature extraction module is used to extract features from the signal within the sliding window based on the results of segmented preprocessing. A wideband oscillation monitoring module is used to predict wideband oscillation states using a fully connected network (FC) based on extracted feature values.

5. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the broadband oscillation monitoring method according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the broadband oscillation monitoring method according to any one of claims 1 to 3.

Images