A waveform extraction-based remote simulation demonstration system and method

Abstract

The application provides a waveform extraction-based remote simulation demonstration system and method, and belongs to the field of remote operation and maintenance testing; solves the problem that the transmission efficiency of the recording wave file is low and the real-time performance is poor when a new power system is tested on site, so that distributed remote simulation testing cannot be realized; the system comprises a real-time digital simulation system, a recording wave device, a control host, a wireless communication network and multiple demonstration devices, wherein the real-time digital simulation system is used for constructing a fault model of a field primary system and converting the fault model into a fault simulation signal output; the recording wave device stores the fault simulation signal output as a recording wave file, extracts fault information based on time domain analysis, extracts electrical quantity characteristic signals of the recording wave file based on multi-algorithm fusion, and forms executable text of a state sequence, so that a numerical simulation test signal is obtained; the control host is used for realizing transmission of the simulation test signal through the wireless communication network according to a mapping relationship between the recording wave file and the demonstration devices; and the application is applied to remote simulation.

Description

Technical Field

[0001] This application relates to the field of remote operation, maintenance and testing of secondary protection in new power systems, and in particular to a remote simulation verification system and method based on waveform extraction. Background Technology

[0002] As a core infrastructure for national energy security, the power system's operational stability, fault response efficiency, and simulation verification capabilities are directly linked. Currently, with the power grid's transformation towards a dual-high ratio of renewable energy and power electronic equipment, and the expansion of inter-regional interconnection, traditional power system simulation and empirical methods face multiple bottlenecks. The large-scale grid connection of renewable energy sources such as wind and solar power results in significant power output fluctuations and randomness; simultaneously, the widespread application of power electronic equipment such as converter valves and inverters leads to frequent complex electromagnetic problems such as grid harmonics and transient impacts. Traditional testing methods often rely on "theoretical models + simplified parameters," making it difficult to accurately reproduce voltage fluctuation waveforms during renewable energy grid connection and current transient waveforms during commutation failures, thus failing to meet the field empirical requirements for coordinated control of power generation, grid, load, and storage.

[0003] Remote simulation field verification based on waveform playback is currently an effective on-site operation and maintenance testing method for new power systems. However, due to the extremely large amount of transient waveform data, cross-regional wireless transmission based on 4G / 5G and other technologies is limited by bandwidth and prone to data loss, resulting in missing simulation data. Therefore, this application proposes a remote simulation verification method based on waveform extraction. This method extracts signal features from waveform recording files using time-domain analysis and wavelet transform, and transmits the relevant information to the field in short message format via wireless communication to achieve remote distributed simulation testing of secondary protection in new power systems. Summary of the Invention

[0004] To address the challenges of insufficient wireless bandwidth and poor regional signal strength in field testing of new power systems, which result in low transmission efficiency and poor real-time performance of waveform recording files and thus hinder distributed remote simulation testing, this application proposes a remote simulation verification system and method based on waveform extraction.

[0005] The technical solution adopted in this application is: a remote simulation verification system based on waveform extraction, including a real-time digital simulation system, a waveform recording device, a control host, a wireless communication network and multiple verification devices. The multiple verification devices are respectively connected to the corresponding protection devices. The real-time digital simulation system is used to construct the fault model of the primary system in the field and output electrical fault simulation signals through a matching power amplifier.

[0006] The waveform recording device acquires the fault simulation signal output by the real-time digital simulation system and stores it as a waveform recording file. The host computer software system built into the waveform recording device extracts fault information based on time domain analysis, extracts electrical quantity feature signals from the waveform recording file based on multi-algorithm fusion, and forms an executable text of state sequence, thereby digitizing the waveform recording file and obtaining digitized simulation test signals.

[0007] The control host is used to transmit the corresponding numerical simulation test signals of the waveform recording files through a wireless communication network, based on the mapping relationship between the waveform recording files and the empirical device.

[0008] The verification device remotely receives the numerical simulation test signal, and in synchronous time synchronization mode, different verification devices remotely restore and output the simulation signal of different field secondary protection devices.

[0009] Furthermore, fault information extraction based on time-domain analysis involves splitting the waveform file into three state sequences corresponding to the pre-fault, during-fault, and post-fault states. Time-domain analysis is then performed on the waveform files corresponding to these states to obtain the fault information for each state sequence, thus digitizing the waveform file.

[0010] Furthermore, the process of extracting electrical quantity feature signals from waveform recordings based on multi-algorithm fusion and forming executable text of state sequences involves three steps:

[0011] 1) Multi-dimensional waveform feature extraction:

[0012] After loading the waveform recording file, waveform features are comprehensively extracted from three dimensions: time domain, frequency domain, and time-frequency domain.

[0013] 2) Signal conversion:

[0014] The waveform features obtained from the recorded waveform file are converted into a data model that dynamically changes based on time-domain information. This data model covers the feature data of both steady-state and transient signals.

[0015] 3) Constructing a mapping mechanism between waveform features and digital twin models:

[0016] A digital twin model is established based on the physical characteristics of the protection device. A gray relational analysis algorithm is used to construct a mapping relationship between waveform features and input parameters of the digital twin model, ensuring that the input of the digital twin model is dynamically synchronized with the actual operating status of the protection device.

[0017] Furthermore, the amplitude and phase of the voltage and current during the fault are extracted using a windowed FFT algorithm;

[0018] The amplitude and phase of voltage and current before and after the fault were extracted using wavelet transform and Kalman filtering.

[0019] Furthermore, multidimensional waveform features include steady-state features, frequency domain features, and transient change features.

[0020] Furthermore, the specific implementation steps of the mapping mechanism between waveform features and digital twin models include:

[0021] (1) Construct a multi-dimensional correlation model of waveform features and model parameters: use Pearson correlation coefficient to calculate the linear correlation between waveform features and digital twin model parameters, capture nonlinear correlation through kernel function method, and establish a bidirectional correlation matrix between waveform features and digital twin model parameters;

[0022] (2) Construct a mapping rule base. The rules include feature identifiers, model parameter identifiers, etc., to realize the association mapping between time domain, frequency domain, and time-frequency domain waveform features and control rules, and finally generate executable text.

[0023] Furthermore, the expression for the data model is as follows:

[0024] ;

[0025] In the formula: For waveform data models, A dc α is the amplitude of the attenuated DC component; t is time; n is the harmonic order; A n denoted as , where is the amplitude of the nth harmonic; B is the DC regulation component; 2πnft is the phase angle of the nth harmonic; and nf is the frequency of the nth harmonic. This represents the initial phase of the nth harmonic.

[0026] Furthermore, the experimental setup uses pulses from the mains power supply for synchronization.

[0027] A remote simulation verification method based on waveform extraction, employing the aforementioned remote simulation verification system based on waveform extraction, includes the following steps:

[0028] S1: Construct a fault model of the primary system in the field through a real-time simulation system and output fault simulation signals;

[0029] S2: The waveform recording device stores the fault simulation signal output by the real-time simulation system as a waveform recording file. The host computer software system built into the waveform recording device extracts fault information and generates executable text of the state sequence to obtain the numerical simulation test signal, and transmits the numerical simulation test signal to the control host.

[0030] S3: The control host realizes the corresponding transmission of numerical simulation test signals through the wireless communication network according to the mapping relationship between the waveform recording file and the empirical device;

[0031] S4: In synchronous time synchronization mode, different verification devices remotely simulate and synchronously restore the output of the field secondary protection device, thereby completing the operation, maintenance and testing of the field secondary protection system in a remote distributed manner.

[0032] Furthermore, the steps to achieve synchronization in step S4 are as follows:

[0033] When the experimental device is powered by 220V AC mains, the acquisition module acquires the voltage waveform of the 220V AC mains power supply and records the PPS pulses sequentially. The PPS pulses acquired from the 220V AC mains voltage waveform are used to synchronize the experimental device with the mains power supply. Different experimental devices track and synchronize electrical quantity signals based on the PPS pulses, ultimately achieving consistent synchronous control output of distributed experimental devices.

[0034] The beneficial effects of this application compared to the prior art are as follows: This application converts the fault simulation waveform of the real-time digital simulation system into state sequence control parameters, and realizes remote control and transmission via wireless network. This solves the problems of insufficient 4G / 5G bandwidth, poor regional signal leading to low transmission efficiency of waveform recording files, and poor real-time performance, which make it impossible to realize distributed remote simulation testing in the field testing of new power systems. It realizes the field simulation and verification of secondary protection of new power systems and improves the reliability of the power grid. Attached Figure Description

[0035] The following description, in conjunction with the accompanying drawings, further illustrates this application:

[0036] Figure 1 The diagram below illustrates the principle of the remote simulation verification method based on waveform extraction, as provided in this embodiment of the application.

[0037] Figure 2 This is a schematic diagram illustrating the principle of extracting feature signals from transient recording files based on multi-algorithm fusion, as given in an embodiment of this application.

[0038] Figure 3 The schematic diagram provided in this application illustrates the synchronization principle of different empirical devices based on the power supply. Detailed Implementation

[0039] like Figures 1 to 3As shown, this application provides a remote simulation verification system based on waveform extraction, including a real-time digital simulation system, a waveform recording device, a control host, a wireless communication network, and multiple verification devices. Each verification device is connected to a corresponding protection device. The real-time digital simulation system is used to construct a fault model of the primary system in the field and outputs simulated electrical fault signals through a matching power amplifier. The waveform recording device acquires the simulated fault signals output by the real-time digital simulation system and stores them as COMTRADE format waveform files. The host computer software system built into the waveform recording device extracts fault information based on time-domain analysis and multi-algorithm fusion. The electrical quantity characteristic signals of the waveform recording file are extracted and formed into an executable text of the state sequence, thereby digitizing the waveform recording file and obtaining the digitized simulation test signal. The control host is used to transmit the corresponding digitized simulation test signal of the waveform recording file through a wireless communication network (4G / 5G network can be used) according to the mapping relationship between the waveform recording file and the demonstration device. The demonstration device remotely receives the digitized simulation test signal, and in synchronous time synchronization mode, different demonstration devices remotely restore and output the simulation signal of different field secondary protection devices, so as to complete the operation, maintenance and testing of the field secondary protection system in a remote distributed manner.

[0040] The steps for extracting fault information based on time-domain analysis include:

[0041] The COMTRADE format waveform file is split into three state sequences: before, during, and after the fault. Time-domain analysis is performed on the waveform files corresponding to these three states to obtain the fault information for each state sequence. This fault information includes the state duration, electrical parameters such as voltage and current, and circuit breaker switch position signals. The electrical parameters include, but are not limited to, characteristic parameters such as the amplitude, frequency, and phase of the voltage and current signals, thereby quantifying the COMTRADE format waveform file.

[0042] The executable text that extracts electrical quantity feature signals from waveform recordings based on multi-algorithm fusion and forms a state sequence includes the following specific implementation steps:

[0043] 1) Multi-dimensional waveform feature extraction:

[0044] After loading the waveform recording file in COMTRADE format, waveform features are comprehensively extracted from three dimensions: time domain, frequency domain, and time-frequency domain. Principal component analysis (PCA) is used to reduce dimensionality and remove redundancy, resulting in a set of core feature parameters that accurately characterize the equipment's operating status. The time domain features reflect the overall amplitude distribution and trend of the waveform, the frequency domain features reveal the frequency composition and amplitude distribution of the waveform, and the time-frequency domain features capture the local features of the waveform at different time-frequency scales. The fusion of these three features achieves a comprehensive characterization of the equipment's operating status.

[0045] This multi-dimensional waveform feature extraction includes:

[0046] (1) Extracting steady-state features based on time-domain algorithms:

[0047] The steady-state mean, variance, and RMS value of the waveform data are calculated using the moving average algorithm.

[0048] The distribution characteristics of steady-state waveforms are identified using a skewness-kurtosis joint algorithm.

[0049] A steady-state feature subset is constructed by extracting steady-state peak values, valley values, and peak-to-peak values ​​using a peak detection algorithm.

[0050] (2) Extracting frequency domain features based on frequency algorithms:

[0051] The fundamental frequency, harmonic order, and amplitude of each harmonic are extracted using Fast Fourier Transform (FFT).

[0052] The frequency domain amplitude distribution characteristics are obtained through the power spectral density estimation (PSD) algorithm;

[0053] The frequency bands are decomposed using wavelet packet transform (WPT), and the energy entropy and information entropy of each frequency band are calculated to construct a frequency domain feature subset.

[0054] (3) Extracting transient change features based on time-frequency domain algorithm:

[0055] A modulus maxima algorithm based on wavelet transform (WT) is used to capture the abrupt change in time and amplitude of transient signals.

[0056] Hilbert-Huang transform (HHT) was used to decompose the waveform data into intrinsic mode functions (IMFs), and the instantaneous frequency, instantaneous amplitude and proportion of each IMF component were extracted.

[0057] The time-frequency joint distribution of the transient signal is obtained by S-transform (ST), the time-frequency clustering features of the abrupt change region are extracted, and a subset of transient change features is constructed.

[0058] 2) Signal conversion:

[0059] The waveform features obtained from the recorded waveform file are converted into a dynamically changing data model based on time-domain information. This data model covers both steady-state and transient signal feature data. The calculation formula for the data model is as follows:

[0060] =Transient model + Steady-state model;

[0061] ;

[0062] In the formula: For waveform data models, A dcα is the amplitude of the attenuated DC component; t is time; n is the harmonic order; A n denoted as , where is the amplitude of the nth harmonic; B is the DC regulation component; 2πnft is the phase angle of the nth harmonic; and nf is the frequency of the nth harmonic. This represents the initial phase of the nth harmonic.

[0063] This data model can be used for signal modulation, waveform playback, and reconstruction.

[0064] 3) Constructing a mapping mechanism between waveform features and digital twin models:

[0065] A digital twin model is established based on the physical characteristics of the protection device (i.e., the device under test). A grey relational analysis algorithm is used to construct a mapping relationship between waveform features and the input parameters of the digital twin model, ensuring dynamic synchronization between the digital twin model input and the actual operating state of the protection device. After initial calibration of the digital twin model, the static values ​​of the actual operating state of the protection device can be adjusted using real-time updated waveform features, making the test output process consistent with the simulation waveform.

[0066] The specific implementation steps of this mapping mechanism include:

[0067] (1) Construct a multi-dimensional correlation model of waveform features and model parameters: use Pearson correlation coefficient to calculate the linear correlation between waveform features and model parameters, capture nonlinear correlation through kernel function method, and establish a bidirectional correlation matrix between waveform features and model parameters;

[0068] (2) Construct a mapping rule base. The rules include feature identifiers, model parameter identifiers, etc., to realize the association mapping between waveform features such as time domain (peak value, mean, variance), frequency domain (fundamental frequency, harmonic content), and time-frequency domain (energy entropy, instantaneous frequency) and control rules, and finally generate executable text.

[0069] Based on the above system, this application also proposes a remote simulation verification method based on waveform extraction, including the following steps:

[0070] S1: Construct a fault model of the primary system in the field through a real-time simulation system and output fault simulation signals;

[0071] S2: The waveform recording device stores the fault simulation signal output by the real-time simulation system as a waveform recording file. The host computer software system built into the waveform recording device extracts fault information and generates executable text of the state sequence to obtain the numerical simulation test signal, and transmits the numerical simulation test signal to the control host.

[0072] S3: The control host realizes the corresponding transmission of numerical simulation test signals through the wireless communication network according to the mapping relationship between the waveform recording file and the empirical device;

[0073] S4: In synchronous time synchronization mode, different verification devices remotely simulate and synchronously restore the output of the field secondary protection device, thereby completing the operation, maintenance and testing of the field secondary protection system in a remote distributed manner.

[0074] In step S2, the waveform recording file is divided into states based on the time-domain spectrum, such as... Figure 2 As shown, the overall duration of the waveform recording file is T. Based on the changes in sampling rate and amplitude during the recording process, the time of the waveform recording file is extracted into three time segments: before the fault (T0), during the fault (T1), and after the fault (T2), where T = T0 + T1 + T2. The sampling rate before and after the fault is 1k, and the sampling rate during the fault is 10k. The voltage amplitude change from before the fault to during the fault is greater than 20%. For the amplitude and phase before and after the fault, wavelet transform and Kalman filtering are used for extraction. For the amplitude and phase during the fault, a windowed FFT algorithm is used for extraction. Finally, a state sequence text covering the amplitude, phase, frequency, and duration of voltage and current is generated.

[0075] In step S4, synchronization of different experimental devices can be achieved based on the power supply, such as... Figure 3 As shown, for distributed empirical devices, a power supply pulse-based approach is used to achieve synchronization among multiple devices. When the empirical devices are powered by 220V AC mains, the acquisition module obtains the 220V AC mains voltage waveform. Synchronization within the 220V AC mains area is achieved using a 50Hz sine wave sampling frequency, i.e., 50 cycles per second, with each cycle lasting 20ms. The time from the first rising edge crossing zero to the fifty-first rising edge crossing zero is 1 second, which can be recorded as one PPS pulse. Subsequent PPS pulses are recorded sequentially. The PPS pulses obtained from the 220V AC mains voltage waveform are used to synchronize the empirical devices with the mains power supply. Different empirical devices track and synchronize electrical quantity signals such as voltage and current based on the PPS pulses, ultimately achieving consistent synchronous control output for the distributed empirical devices.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A remote simulation verification system based on waveform extraction, characterized in that: It includes a real-time digital simulation system, a waveform recording device, a control host, a wireless communication network, and multiple verification devices. The multiple verification devices are connected to the corresponding protection devices. The real-time digital simulation system is used to construct a fault model of the primary system in the field and outputs electrical fault simulation signals through a matching power amplifier. The waveform recording device acquires the fault simulation signal output by the real-time digital simulation system and stores it as a waveform recording file. The host computer software system built into the waveform recording device extracts fault information based on time domain analysis, extracts electrical quantity feature signals from the waveform recording file based on multi-algorithm fusion, and forms an executable text of state sequence, thereby digitizing the waveform recording file and obtaining digitized simulation test signals. The control host is used to transmit the corresponding numerical simulation test signals of the waveform recording files through a wireless communication network, based on the mapping relationship between the waveform recording files and the empirical device. The verification device remotely receives the numerical simulation test signal, and in synchronous time synchronization mode, different verification devices remotely restore and output the simulation signal of different field secondary protection devices.

2. The remote simulation verification system based on waveform extraction according to claim 1, characterized in that: Fault information extraction based on time-domain analysis involves splitting the waveform file into three state sequences: before, during, and after the fault. Time-domain analysis is then performed on the waveform files corresponding to these three states to obtain the fault information for each state sequence, thus digitizing the waveform file.

3. The remote simulation verification system based on waveform extraction according to claim 1, characterized in that: The executable text that extracts electrical quantity feature signals from waveform recordings and forms state sequences based on multi-algorithm fusion includes three steps: 1) Multi-dimensional waveform feature extraction: After loading the waveform recording file, waveform features are comprehensively extracted from three dimensions: time domain, frequency domain, and time-frequency domain. 2) Signal conversion: The waveform features obtained from the recorded waveform file are converted into a data model that dynamically changes based on time-domain information. This data model covers the feature data of both steady-state and transient signals. 3) Constructing a mapping mechanism between waveform features and digital twin models: A digital twin model is established based on the physical characteristics of the protection device. A gray relational analysis algorithm is used to construct a mapping relationship between waveform features and input parameters of the digital twin model, ensuring that the input of the digital twin model is dynamically synchronized with the actual operating status of the protection device.

4. The remote simulation verification system based on waveform extraction according to claim 2, characterized in that: The magnitude and phase of the voltage and current during the fault were extracted using a windowed FFT algorithm. The amplitude and phase of voltage and current before and after the fault were extracted using wavelet transform and Kalman filtering.

5. The remote simulation verification system based on waveform extraction according to claim 3, characterized in that: Multidimensional waveform features include steady-state features, frequency domain features, and transient change features.

6. The remote simulation verification system based on waveform extraction according to claim 3, characterized in that: The specific implementation steps of the mapping mechanism between waveform features and digital twin models include: (1) Construct a multi-dimensional correlation model of waveform features and model parameters: use Pearson correlation coefficient to calculate the linear correlation between waveform features and digital twin model parameters, capture nonlinear correlation through kernel function method, and establish a bidirectional correlation matrix between waveform features and digital twin model parameters; (2) Construct a mapping rule base. The rules include feature identifiers, model parameter identifiers, etc., to realize the association mapping between time domain, frequency domain, and time-frequency domain waveform features and control rules, and finally generate executable text.

7. The remote simulation verification system based on waveform extraction according to claim 3, characterized in that: The data model expression is as follows: ； In the formula: For waveform data models, A dc α is the amplitude of the attenuated DC component; t is the attenuation coefficient; t is time. n is the harmonic order; A n denoted as , where is the amplitude of the nth harmonic; B is the DC regulation component; 2πnft is the phase angle of the nth harmonic; and nf is the frequency of the nth harmonic. This represents the initial phase of the nth harmonic.

8. A remote simulation verification system based on waveform extraction according to claim 3, characterized in that: The experimental setup uses pulses from the mains power supply for synchronization.

9. A remote simulation verification method based on waveform extraction, characterized in that: The remote simulation verification system based on waveform extraction as described in any one of claims 1-8 includes the following steps: S1: Construct a fault model of the primary system in the field through a real-time simulation system and output fault simulation signals; S2: The waveform recording device stores the fault simulation signal output by the real-time simulation system as a waveform recording file. The host computer software system built into the waveform recording device extracts fault information and generates executable text of the state sequence to obtain the numerical simulation test signal, and transmits the numerical simulation test signal to the control host. S3: The control host realizes the corresponding transmission of numerical simulation test signals through the wireless communication network according to the mapping relationship between the waveform recording file and the empirical device; S4: In synchronous time synchronization mode, different verification devices remotely simulate and synchronously restore the output of the field secondary protection device, thereby completing the operation, maintenance and testing of the field secondary protection system in a remote distributed manner.

10. The remote simulation verification method based on waveform extraction according to claim 9, characterized in that: The steps to achieve synchronization in step S4 are as follows: When the experimental device is powered by 220V AC mains, the acquisition module acquires the voltage waveform of the 220V AC mains power supply and records the PPS pulses sequentially. The PPS pulses acquired from the 220V AC mains voltage waveform are used to synchronize the experimental device with the mains power supply. Different experimental devices track and synchronize electrical quantity signals based on the PPS pulses, ultimately achieving consistent synchronous control output of distributed experimental devices.

Images