Flexible DC converter valve DC support capacitor current on-line monitoring method and system

Abstract

The invention discloses a flexible DC converter valve DC support capacitor current on-line monitoring method and system, and the method comprises the steps: controlling a Rogowski coil to collect the operation data of a selected capacitor in a flexible DC converter valve in response to an external sampling starting instruction; sequentially performing amplification, filtering and abnormal value detection processing on the current waveform data to obtain to-be-corrected current data; based on a pre-constructed neural network model, performing correction processing on the to-be-corrected current data to obtain to-be-analyzed current data; analyzing the to-be-analyzed current data based on a big data analysis technology and a machine learning algorithm; and according to an analysis result, a generated current regulation and control instruction matched with the analysis result is sent to the selected capacitor. According to the method provided by the invention, the monitoring precision of the capacitor is improved.

Description

Technical Field

[0001] This application relates to the technical field of power monitoring, and particularly to an online monitoring method for the current of the DC support capacitor of a flexible DC converter valve. Background Art

[0002] In the power system, the DC support capacitor is one of the core components in the flexible DC converter valve. As the core equipment of the flexible DC transmission project, the converter valve is responsible for realizing the conversion between alternating current and direct current and controlling parameters such as voltage and current. The DC support capacitor plays a key supporting role in this process. As an important reactive power compensation device, the DC support capacitor plays a crucial role in improving the power factor of the power grid, reducing line losses, and improving voltage quality.

[0003] However, the current change is instantaneous. The traditional capacitor monitoring methods mainly rely on manual inspection and regular detection. This method cannot accurately measure the real-time current change in the capacitor. Especially in the case of small current or fast-changing current, the lack of accuracy may lead to misjudgment of the capacitor state, thus affecting the stability and safety of the flexible DC conversion system. Summary of the Invention

[0004] This application provides an online monitoring method and system for the current of the DC support capacitor of a flexible DC converter valve to solve the technical problem of how to improve the accuracy of capacitor current monitoring.

[0005] To solve the above technical problem, an embodiment of this application provides an online monitoring method for the current of the DC support capacitor of a flexible DC converter valve, including:

[0006] In response to an external sampling start instruction, control a Rogowski coil to collect the operation data of a selected capacitor in the flexible DC converter valve, where the operation data at least includes the current waveform data of the connection terminal posts of the selected capacitor;

[0007] Perform amplification, filtering, and outlier detection processing on the current waveform data in sequence to obtain the current data to be corrected;

[0008] Based on a pre-constructed neural network model, perform correction processing on the current data to be corrected to obtain the current data to be analyzed, where the correction processing is configured to eliminate the current amplitude-frequency characteristic fluctuation caused by instantaneous current disorder;

[0009] Analyze the current data to be analyzed based on big data analysis technology and machine learning algorithms;

[0010] According to the analysis result, send the generated current regulation instruction that matches it to the selected capacitor.

[0011] As one of the preferred solutions, in response to an external sampling start instruction, controlling the Rogowski coil to collect the operation data of a selected capacitor in the flexible DC converter valve, including:

[0012] Collecting the charging and discharging current of the selected capacitor based on the Rogowski coil and converting the charging and discharging current into a proportional induced electromotive force;

[0013] Performing digital conversion on the induced electromotive force to obtain a digital signal;

[0014] Sampling the digital signal to obtain current waveform data.

[0015] As one of the preferred solutions, sequentially performing amplification, filtering, and outlier detection processing on the current waveform data to obtain current data to be corrected, including:

[0016] Inputting the current waveform data into an operational amplifier and adjusting the feedback resistance of the operational amplifier to amplify the current waveform data to a preset multiple;

[0017] Inputting the amplified current waveform data into a filter for filtering processing and outputting a filtered signal;

[0018] Based on a statistical anomaly detection method, performing anomaly identification on the filtered current waveform data to obtain outliers, and performing deletion, replacement, or marking processing on the outliers to obtain the current data to be corrected.

[0019] As one of the preferred solutions, based on a pre-constructed neural network model, performing correction processing on the current data to be corrected to obtain current data to be analyzed, including:

[0020] Obtaining historical current waveform data, where the historical current waveform data includes normal current waveform data and abnormal current waveform data;

[0021] Constructing an initial neural network model for correction and training the initial neural network model based on the historical current waveform data to obtain a trained neural network model;

[0022] Inputting the current data to be corrected into the neural network model for processing to obtain corrected current data to be analyzed.

[0023] As one of the preferred solutions, analyzing the current data to be analyzed based on big data analysis technology and machine learning algorithms, including:

[0024] Extracting waveform feature data of the current waveform data, where the waveform feature data includes peak data, valley data, average value data, and fluctuation range data;

[0025] Construct an initial waveform recognition model, and train the initial waveform recognition model by using the acquired historical waveform data and the corresponding anomaly labels to obtain a trained waveform recognition model;

[0026] Input the waveform features into the waveform recognition model to classify the waveform features, and obtain the working state of the selected capacitor.

[0027] Another embodiment of the present application provides an online monitoring system for the current of a DC support capacitor of a flexible DC converter valve, which is characterized by including:

[0028] An acquisition module, configured to control a Rogowski coil to acquire the operation data of a selected capacitor in the flexible DC converter valve in response to an external sampling start instruction, where the operation data at least includes the current waveform data of the connection terminal posts of the selected capacitor;

[0029] A preprocessing module, configured to sequentially perform amplification, filtering, and outlier detection processing on the current waveform data to obtain current data to be corrected;

[0030] A correction module, configured to perform correction processing on the current data to be corrected based on a pre-constructed neural network model to obtain current data to be analyzed, where the correction processing is configured to eliminate the current amplitude-frequency characteristic fluctuations caused by instantaneous current disorders;

[0031] An analysis module, configured to analyze the current data to be analyzed based on big data analysis technology and machine learning algorithms;

[0032] A monitoring module, configured to send a generated current regulation instruction matching the analysis result to the selected capacitor.

[0033] As one of the preferred solutions, the acquisition module is specifically configured to:

[0034] Collect the charging and discharging current of the selected capacitor based on the Rogowski coil, and convert the charging and discharging current into a proportional induced electromotive force;

[0035] Perform digital conversion on the induced electromotive force to obtain a digital signal;

[0036] Sample the digital signal to obtain current waveform data.

[0037] As one of the preferred solutions, the preprocessing module is specifically configured to:

[0038] Input the current waveform data into an operational amplifier, and adjust the feedback resistance of the operational amplifier to amplify the current waveform data to a preset multiple;

[0039] Input the amplified current waveform data into a filter for filtering, and output the filtered signal;

[0040] Based on a statistical anomaly detection method, identify anomalies in the filtered current waveform data to obtain anomaly values, and perform deletion, replacement, or marking on the anomaly values to obtain the current data to be corrected.

[0041] As one of the preferred solutions, the correction module is specifically configured to:

[0042] Obtain historical current waveform data, where the historical current waveform data includes normal current waveform data and abnormal current waveform data;

[0043] Construct an initial neural network model for correction, and train the initial neural network model based on the historical current waveform data to obtain a trained neural network model;

[0044] Input the current data to be corrected into the neural network model for processing to obtain the current data to be analyzed after correction.

[0045] As one of the preferred solutions, the analysis module is specifically configured to:

[0046] Extract the waveform feature data of the current waveform data, where the waveform feature data includes peak data, valley data, average value data, and fluctuation range data;

[0047] Construct an initial waveform recognition model, and train the initial waveform recognition model through the obtained historical waveform data and corresponding anomaly labels to obtain a trained waveform recognition model;

[0048] Input the waveform features into the waveform recognition model to classify the waveform features and obtain the operating state of the selected capacitor.

[0049] Compared with the prior art, the beneficial effects of the embodiments of the present application are at least one of the following:

[0050] 1) In the present application, the operating data of the selected capacitor in the flexible DC converter valve, especially the current waveform data of the terminal post, is collected through a Rogowski coil. This method has high sensitivity and accuracy and can capture the subtle characteristics of current changes.

[0051] 2) In the present application, the collected current waveform data is amplified, filtered, and anomaly value detection is performed, which can effectively remove noise and interference, improve the signal-to-noise ratio of the data, and provide a high-quality data basis for subsequent analysis.

[0052] 3) This application uses a pre - constructed neural network model to correct the current data to be corrected, which can eliminate the fluctuations in the current amplitude - frequency characteristics caused by instantaneous current disorders, and further improve the accuracy and reliability of the data. At the same time, this application analyzes the processed current data based on big data analysis technology, which can discover the potential laws and characteristics in the data and provide strong support for fault prediction and diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a schematic flowchart of the online monitoring method for the current of the DC support capacitor of the flexible DC converter valve in one embodiment of this application;

[0054] Figure 2 is a schematic diagram of the current acquisition device in one embodiment of this application;

[0055] Figure 3 is a schematic diagram of the online monitoring system for the current of the DC support capacitor of the flexible DC converter valve in one embodiment of this application;

[0056] REFERENCE SIGNS:

[0057] Among them, 1, acquisition circuit board; 2, upper cover plate; 3, elastic fastening ring; 4, insulating housing; 5, temperature sensing device; 6, energy - harvesting acquisition coil. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. The purpose of providing these embodiments is to make the disclosure content of this application more thorough and comprehensive. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0059] In the description of this application, the terms "first", "second", "third", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first", "second", "third", etc. may explicitly or implicitly include one or more of such features. In the description of this application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0060] In the description of the present application, it should be noted that unless otherwise clearly defined and limited, the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower", and similar expressions used herein are only for the purpose of illustration, rather than indicating or implying that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be construed as a limitation to the present application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0061] In the description of the present application, it should be noted that unless otherwise defined, all technical and scientific terms used in the present application have the same meaning as those commonly understood by those skilled in the technical field to which the present application belongs. The terms used in the specification of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0062] An embodiment of the present application provides a capacitor monitoring system. Specifically, please refer to Figure 1 , Figure 1 which shows a schematic flow chart of the online monitoring method for the current of the DC support capacitor of the flexible DC converter valve in one of the embodiments of the present application, and it includes steps S1 - S5:

[0063] S1: In response to an external sampling start instruction, control the Rogowski coil to collect the operation data of the selected capacitor in the flexible DC converter valve, wherein the operation data at least includes the current waveform data of the connection terminal posts of the selected capacitor;

[0064] According to Faraday's law of electromagnetic induction, when current passes through a conductor, a magnetic field will be generated around the conductor; conversely, when the magnetic field changes, an induced electromotive force will also be generated in the conductor. The Rogowski coil works based on this principle. When the capacitor charges and discharges, the change in the current at its connection terminal posts will generate an induced electromotive force in the Rogowski coil. This induced electromotive force is proportional to the current at the connection terminal posts of the capacitor, and therefore can be used to reflect the change in current.

[0065] Preferably, in an embodiment of the present application, the step of "in response to an external sampling start instruction, control the Rogowski coil to collect the operation data of the selected capacitor in the flexible DC converter valve" includes:

[0066] Collect the charging and discharging current of the selected capacitor based on a Rogowski coil, and convert the charging and discharging current into a proportional induced electromotive force;

[0067] Perform digital conversion on the induced electromotive force to obtain a digital signal;

[0068] Sample the digital signal to obtain current waveform data.

[0069] For the convenience of subsequent data processing and analysis, it is necessary to convert the induced electromotive force into a digital signal. This process is usually achieved through an analog-to-digital converter (ADC). The analog-to-digital converter can convert a continuous analog signal (such as the induced electromotive force) into a discrete digital signal. In this application, the analog-to-digital converter converts the induced electromotive force output by the Rogowski coil into a digital signal for subsequent sampling and processing.

[0070] Sampling is one of the key steps in digital signal acquisition. In this application, a sampler samples the digital signal output by the analog-to-digital converter to obtain a series of discrete sampling points. These sampling points reflect the change of the current at the capacitor terminal post, that is, the current waveform data. The selection of the sampling frequency should satisfy the Nyquist sampling theorem to avoid aliasing.

[0071] Specifically, refer to Figure 2 , Figure 2 which is a schematic diagram of a current acquisition device provided by an embodiment of this application. Among them, the insulating housing is the basic structure of the current acquisition device, made of high-quality insulating materials to ensure that the internal circuit is isolated from the external environment, preventing moisture, dust, and corrosive substances from invading, thereby protecting the internal circuit from damage. The upper cover plate is responsible for enclosing the current acquisition device to protect the internal circuit from physical damage. It is usually made of strong and durable materials, with sufficient strength and stiffness to withstand external impacts and vibrations. The outer shell size of the sensing unit is adaptively designed according to the size of the capacitor terminal insulator to ensure that it can be closely mounted at the bottom of the insulator. This design not only improves the measurement accuracy but also ensures the stability of the sensing unit under various working conditions.

[0072] The Rogowski coil is a non-contact current sensor that can measure the charging and discharging current of a capacitor without disconnecting the circuit. Its working principle is based on Faraday's law of electromagnetic induction. When current passes through the coil, an induced electromotive force proportional to the current is generated. The number of turns and size of the Rogowski coil are designed according to the current measurement range and accuracy requirements. The more turns, the greater the induced electromotive force, but it will also increase the volume and weight of the coil. Therefore, on the premise of meeting the measurement requirements, the number of turns should be minimized as much as possible to optimize the structure of the coil. At the same time, in order to convert the induced electromotive force into a voltage signal, two 1Ω high-precision resistors are connected in parallel on the secondary side of the Rogowski coil in this application. This design method can not only improve the signal-to-noise ratio of the signal, but also effectively reduce the measurement error.

[0073] The elastic fastening ring ensures that the sensing unit can be fastened to the bottom of the capacitor terminal insulator, and at the same time provides a certain deformation ability to adapt to insulators of different sizes. This design ensures that the sensing unit can remain stable under various working conditions.

[0074] The acquisition circuit board is the core part of the current acquisition device, integrating circuits such as current acquisition, signal conditioning, and communication. These circuits work together to realize the digital acquisition and processing of the charging and discharging current waveform of the capacitor and the temperature of the bottom shell of the terminal. In an embodiment of the present application, the acquisition circuit board includes an STM32 single-chip microcomputer.

[0075] S2: Sequentially perform amplification, filtering, and outlier detection processing on the current waveform data to obtain the current data to be corrected;

[0076] Specifically, when the capacitor is charging and discharging, current passes through the energy-taking acquisition coil (Rogowski coil), and an induced electromotive force proportional to the current is generated according to Faraday's law of electromagnetic induction. This induced electromotive force is then amplified and filtered by the signal conditioning circuit on the acquisition circuit board to remove noise and interference components. The processed current signal is converted into a digital signal for subsequent analysis. The acquired current waveform data is processed by the microprocessor on the acquisition circuit board, including steps such as data compression and formatting, to reduce the bandwidth of data transmission and the occupancy of storage space. Subsequently, these data are transmitted to the acquisition control unit through a wireless communication module or a fiber optic communication interface for further analysis and processing.

[0077] Preferably, in an embodiment of the present application, the sequentially performing amplification, filtering, and outlier detection processing on the current waveform data to obtain the current data to be corrected includes:

[0078] Input the current waveform data into an operational amplifier, and by adjusting the feedback resistor of the operational amplifier, amplify the current waveform data to a preset multiple;

[0079] Input the amplified current waveform data into a filter for filtering processing to output a filtered signal;

[0080] Perform anomaly recognition on the filtered current waveform data based on a statistical anomaly detection method to obtain anomaly values, and perform deletion, replacement, or marking processing on the anomaly values to obtain the current data to be corrected.

[0081] In this step, an operational amplifier AD627 is used to construct an amplification circuit. AD627 is a high-precision, low-noise operational amplifier suitable for amplifying weak signals. The amplification factor is adjusted according to the signal strength and processing requirements. If the signal is weak, a higher amplification factor is required; if the signal is strong, the amplification factor can be appropriately reduced to avoid signal distortion. By adjusting the resistance values of the feedback resistor and input resistor in the amplification circuit, precise control of the amplification factor can be achieved.

[0082] Furthermore, the current signal may contain high-frequency noise and interference components, which will affect the accuracy and stability of the signal. Therefore, it is necessary to filter through a filter circuit to improve the signal quality. In this module, an RC filter circuit is used to filter the amplified voltage signal. The design of the filter circuit needs to consider the balance between the frequency characteristics of the signal and the filtering effect. If the filtering effect is too strong, useful signal components may be filtered out; if the filtering effect is too weak, noise and interference components cannot be effectively removed. Therefore, appropriate filter elements and parameters need to be selected according to the frequency characteristics of the signal and processing requirements.

[0083] The voltage signal after amplification and filtering needs to be converted into a digital signal for subsequent digital signal processing and analysis. In this module, a high-precision analog-to-digital converter (ADC) is used to sample and quantize the voltage signal. The sampling rate and resolution of the ADC are selected according to the signal characteristics and processing requirements. The higher the sampling rate, the more signal details can be captured; the higher the resolution, the higher the signal accuracy that can be represented. The digitized signal can be further processed such as filtering, smoothing, and feature extraction to more accurately reflect the current state of the capacitor. These processing results can provide strong support for subsequent fault diagnosis, early warning, and condition assessment.

[0084] In the current waveform data upload mode, the STM32 single-chip microcomputer samples the current signal at a frequency of 50 kHz through the internal ADC for 20 ms, obtaining a total of 1000 data points. These waveform data are divided into 40 packets, with each packet containing approximately 25 data points. Then, the 40 packets of data are sequentially transmitted to the data aggregation unit through the Lora communication sub-unit of the data receiving unit. The transmission time of each packet of data is approximately 75 ms, and the total transmission time is approximately 3 seconds.

[0085] S3: Based on a pre - constructed neural network model, perform correction processing on the to - be - corrected current data to obtain to - be - analyzed current data, where the correction processing is configured to eliminate the current amplitude - frequency characteristic fluctuations caused by current instantaneous disorders;

[0086] Preferably, in an embodiment of the present application, the performing correction processing on the to - be - corrected current data based on a pre - constructed neural network model to obtain to - be - analyzed current data includes:

[0087] Obtain historical current waveform data, where the historical current waveform data includes normal current waveform data and abnormal current waveform data;

[0088] Construct an initial neural network model for correction, and train the initial neural network model based on the historical current waveform data to obtain a trained neural network model;

[0089] Input the to - be - corrected current data into the neural network model for processing to obtain the corrected to - be - analyzed current data.

[0090] In this step, historical data containing normal current waveform data and abnormal current waveform data is collected. These data should cover various current change scenarios, including current fluctuations under normal operating conditions and abnormal current fluctuations caused by instantaneous disorders, with the aim of ensuring the diversity and representativeness of the data, so as to be able to train a neural network model with strong generalization ability.

[0091] Select a neural network model suitable for processing time - series data, such as a recurrent neural network (RNN) or its variants (such as LSTM, GRU), or a convolutional neural network (CNN) combined with a one - dimensional convolutional layer to process the current waveform data. Use the historical current waveform data to train the initial neural network model. During the training process, the model will learn how to identify and correct abnormal fluctuations in the current data. Through optimization methods such as backpropagation algorithm and gradient descent, continuously adjust the weight and bias parameters of the model to minimize the prediction error.

[0092] Perform necessary post - processing on the corrected current waveform data, such as anti - normalization, etc., to restore the original dimension of the data. Use a validation dataset (if available) to verify the accuracy and reliability of the corrected current waveform data. The correction effect can be evaluated by calculating error metrics before and after correction (such as mean square error MSE, correlation coefficient, etc.).

[0093] S4: Analyze the to - be - analyzed current data based on big data analysis technology and machine learning algorithms;

[0094] Preferably, in an embodiment of the present application, analyzing the current data to be analyzed based on big data analysis technology and machine learning algorithms includes:

[0095] Extracting waveform feature data of the current waveform data, where the waveform feature data includes peak data, valley data, average value data, and fluctuation range data;

[0096] Constructing an initial waveform recognition model, and training the initial waveform recognition model with the obtained historical waveform data and corresponding anomaly labels to obtain a trained waveform recognition model;

[0097] Inputting the waveform features into the waveform recognition model to classify the waveform features, and obtaining the working state of the selected capacitor.

[0098] Specifically, obtain the original current waveform data from data sources (such as sensors, data acquisition systems, etc.). These data are usually presented in the form of time series, containing information about the change of current over time. Process the original current waveform data to extract key waveform feature data. These feature data are used for subsequent analysis and recognition. Common waveform features include peak data (the maximum value in the current waveform), valley data (the minimum value in the current waveform), average value data (the arithmetic average of the current waveform), and fluctuation range data (the difference between the maximum and minimum values in the current waveform).

[0099] According to the characteristics of the current waveform data and the analysis requirements, select a suitable machine learning algorithm to construct an initial waveform recognition model. Commonly used machine learning algorithms include decision trees, support vector machines (SVMs), neural networks, etc. These algorithms can output classification results based on the input feature data (such as peaks, valleys, average values, and fluctuation ranges).

[0100] Use the obtained historical waveform data and corresponding anomaly labels to train the initial waveform recognition model. The historical waveform data should include normal and abnormal current waveform samples, and the anomaly labels are used to indicate whether each sample belongs to the abnormal category. During the training process, the model will learn how to identify the category (normal or abnormal) of the current waveform based on the input feature data. By adjusting the parameters and structure of the model, the performance of the model can be optimized, and the accuracy and generalization ability of the classification can be improved.

[0101] Input the extracted waveform feature data (peaks, valleys, average values, and fluctuation ranges) into the trained waveform recognition model. The waveform recognition model will classify the current waveform based on the input feature data and output the classification result (normal or abnormal). If the classification result is abnormal, it may be necessary to further analyze the cause and degree of the abnormality in order to take corresponding measures.

[0102] In the embodiments of the present application, the classification result can also be used to judge the working state of the selected capacitor. If the current waveform is recognized as abnormal, it may mean that there are problems such as capacitor failure or performance degradation. By regularly or real-time monitoring the current waveform data and analyzing it, potential problems of the capacitor can be discovered in time, thus avoiding the occurrence of failures or reducing the impact of failures.

[0103] S5: According to the analysis result, send the generated current regulation instruction that matches it to the selected capacitor.

[0104] Specifically, interpret the analysis result of the current data to clarify the characteristics, abnormal points and possible reasons of the current waveform. Analyze the response of the capacitor to current fluctuations to judge whether there are performance degradation, failures or parameters that need to be adjusted. According to the analysis result, determine the current parameters that need to be adjusted, such as current magnitude, frequency, phase, etc. Design the generation logic of the current regulation instruction to ensure that the instruction can accurately reflect the requirements of the analysis result. Check the generated current regulation instruction to ensure that it meets the control requirements and safety specifications of the capacitor. Optimize the instruction according to the actual situation to improve the accuracy and efficiency of regulation. Send the generated current regulation instruction to the selected capacitor through an appropriate communication protocol and interface. Ensure that the instruction is not interfered with and lost during transmission, and maintain its integrity and accuracy. After the capacitor receives the instruction, adjust its working state according to the requirements of the instruction. Monitor the current change of the capacitor during the adjustment process to ensure that it is adjusted according to the instruction requirements.

[0105] Another embodiment of the present application provides an online current monitoring system for the DC support capacitor of a flexible DC converter valve. Specifically, please refer to Figure 3 , Figure 3 which shows a schematic diagram of the online current monitoring system for the DC support capacitor of the flexible DC converter valve in one of the embodiments of the present application. It includes: a collection module 11, a preprocessing module 12, a calibration module 13, an analysis module 14, a monitoring module 15,

[0106] The collection module 11 is used to control the Rogowski coil to collect the operation data of the selected capacitor in the flexible DC converter valve in response to an external sampling start instruction, where the operation data at least includes the current waveform data of the connection terminal posts of the selected capacitor;

[0107] The preprocessing module 12 is used to sequentially perform amplification, filtering and outlier detection processing on the current waveform data to obtain the current data to be calibrated;

[0108] The calibration module 13 is used to perform calibration processing on the current data to be calibrated based on a pre-constructed neural network model to obtain the current data to be analyzed, where the calibration processing is configured to eliminate the current amplitude-frequency characteristic fluctuations caused by instantaneous current disorders;

[0109] An analysis module 14, configured to analyze the to-be-analyzed current data based on big data analysis techniques and machine learning algorithms;

[0110] A monitoring module 15, configured to send the generated current regulation instruction that matches the analysis result to the selected capacitor.

[0111] Preferably, in an embodiment of the present application, the acquisition module is specifically configured to:

[0112] Collect the charge and discharge current of the selected capacitor based on a Rogowski coil, and convert the charge and discharge current into a proportional induced electromotive force;

[0113] Perform digital conversion on the induced electromotive force to obtain a digital signal;

[0114] Sample the digital signal to obtain current waveform data.

[0115] Preferably, in an embodiment of the present application, the preprocessing module is specifically configured to:

[0116] Input the current waveform data into an operational amplifier, and adjust the feedback resistance of the operational amplifier to amplify the current waveform data to a preset multiple;

[0117] Input the amplified current waveform data into a filter for filtering, and output a filtered signal;

[0118] Perform anomaly recognition on the filtered current waveform data based on a statistical anomaly detection method to obtain anomaly values, and perform deletion, replacement, or marking processing on the anomaly values to obtain the to-be-corrected current data.

[0119] Preferably, in an embodiment of the present application, the correction module is specifically configured to:

[0120] Obtain historical current waveform data, where the historical current waveform data includes normal current waveform data and abnormal current waveform data;

[0121] Construct an initial neural network model for correction, and train the initial neural network model based on the historical current waveform data to obtain a trained neural network model;

[0122] Input the to-be-corrected current data into the neural network model for processing to obtain the corrected to-be-analyzed current data.

[0123] Preferably, in an embodiment of the present application, the analysis module is specifically configured to:

[0124] Extracting waveform characteristic data of the current waveform data, wherein the waveform characteristic data includes peak value data, valley value data, average value data and fluctuation range data;

[0125] Constructing an initial waveform recognition model, and training the initial waveform recognition model by using the acquired historical waveform data and corresponding abnormal labels to obtain a trained waveform recognition model;

[0126] The waveform features are input into the waveform recognition model to classify the waveform features, thereby obtaining the working state of the selected capacitor.

[0127] Compared with the prior art, the beneficial effects of the embodiments of the present application are at least one of the following:

[0128] 1) The present application collects the operating data of the selected capacitor in the flexible DC converter valve through the Rogowski coil, especially the current waveform data of the terminal. This method has high sensitivity and accuracy and can capture the subtle characteristics of current changes.

[0129] 2) This application amplifies, filters and detects outliers on the collected current waveform data, which can effectively remove noise and interference, improve the signal-to-noise ratio of the data, and provide a high-quality data basis for subsequent analysis.

[0130] 3) This application uses a pre-built neural network model to correct the current data to be corrected, which can eliminate the fluctuation of the current amplitude-frequency characteristics caused by the instantaneous current disorder, and further improve the accuracy and reliability of the data. At the same time, this application analyzes the processed current data based on big data analysis technology, which can dig out the potential laws and characteristics in the data and provide strong support for fault prediction and diagnosis.

[0131] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An on-line monitoring method for the current of the DC support capacitor of a flexible HVDC converter valve, characterized in that, Comprising: In response to an external sampling start instruction, controlling a Rogowski coil to collect operation data of a selected capacitor in a flexible DC converter valve, wherein the operation data at least includes current waveform data of the connection terminal posts of the selected capacitor; Successively performing amplification, filtering, and outlier detection processing on the current waveform data to obtain current data to be corrected; Based on a pre-constructed neural network model, performing correction processing on the current data to be corrected to obtain current data to be analyzed, wherein the correction processing is configured to eliminate fluctuations in the current amplitude-frequency characteristic caused by instantaneous current disorders; Analyzing the current data to be analyzed based on big data analysis techniques and machine learning algorithms; According to the analysis result, sending a generated current regulation instruction matching thereto to the selected capacitor.

2. The online monitoring method for the current of the DC support capacitor of the flexible DC converter valve according to claim 1, characterized in that, The controlling a Rogowski coil to collect operation data of a selected capacitor in a flexible DC converter valve in response to an external sampling start instruction includes: Collecting the charging and discharging current of the selected capacitor based on the Rogowski coil and converting the charging and discharging current into a proportional induced electromotive force; Performing digital conversion on the induced electromotive force to obtain a digital signal; Sampling the digital signal to obtain current waveform data.

3. The online monitoring method for the current of the DC support capacitor of the flexible DC converter valve according to claim 1, characterized in that, The successively performing amplification, filtering, and outlier detection processing on the current waveform data to obtain current data to be corrected includes: Inputting the current waveform data into an operational amplifier and adjusting the feedback resistance of the operational amplifier to amplify the current waveform data to a preset multiple; Inputting the amplified current waveform data into a filter for filtering processing and outputting a filtered signal; Based on a statistical anomaly detection method, performing anomaly identification on the current waveform data after filtering processing to obtain outliers, and performing deletion, replacement, or marking processing on the outliers to obtain the current data to be corrected.

4. The on-line monitoring method for the current of the DC support capacitor of the flexible DC converter valve according to claim 1, characterized in that The performing correction processing on the current data to be corrected based on a pre-constructed neural network model to obtain current data to be analyzed includes: Obtaining historical current waveform data, wherein the historical current waveform data includes normal current waveform data and abnormal current waveform data; Constructing an initial neural network model for correction, and training the initial neural network model based on the historical current waveform data to obtain a trained neural network model; Inputting the current data to be corrected into the neural network model for processing to obtain corrected current data to be analyzed.

5. The on-line monitoring method for the current of the DC support capacitor of the flexible DC converter valve according to claim 1, characterized in that, The analyzing the current data to be analyzed based on big data analysis techniques and machine learning algorithms includes: Extracting waveform feature data of the current waveform data, wherein the waveform feature data includes peak value data, valley value data, average value data, and fluctuation range data; Constructing an initial waveform recognition model, and training the initial waveform recognition model through the obtained historical waveform data and corresponding anomaly labels to obtain a trained waveform recognition model; Inputting the waveform features into the waveform recognition model to classify the waveform features to obtain the working state of the selected capacitor.

6. An online monitoring system for the current of the DC support capacitor of a flexible DC converter valve, characterized in that, Comprising: The acquisition module is used to control the Rogowski coil to collect the operation data of the selected capacitor in the flexible DC converter valve in response to an external sampling start instruction, where the operation data at least includes the current waveform data of the connection terminal posts of the selected capacitor; The preprocessing module is used to perform amplification, filtering, and outlier detection processing on the current waveform data in sequence to obtain the current data to be corrected; The correction module is used to perform correction processing on the current data to be corrected based on a pre-constructed neural network model to obtain the current data to be analyzed, where the correction processing is configured to eliminate the current amplitude-frequency characteristic fluctuations caused by instantaneous current disorders; The analysis module is used to analyze the current data to be analyzed based on big data analysis technology and machine learning algorithms; The monitoring module is used to send the generated current regulation instruction that matches the analysis result to the selected capacitor.

7. The online monitoring system for the current of the DC support capacitor of the flexible DC converter valve according to claim 6, characterized in that The acquisition module is specifically used for: Collect the charging and discharging current of the selected capacitor based on the Rogowski coil and convert the charging and discharging current into a proportional induced electromotive force; Perform digital conversion on the induced electromotive force to obtain a digital signal; Sample the digital signal to obtain current waveform data.

8. The online monitoring system for the current of the DC support capacitor of the flexible DC converter valve according to claim 6, characterized in that, The preprocessing module is specifically used for: Input the current waveform data into an operational amplifier, and adjust the feedback resistor of the operational amplifier to amplify the current waveform data to a preset multiple; Input the amplified current waveform data into a filter for filtering processing, and output the filtered signal; Perform outlier recognition on the current waveform data after filtering processing based on a statistical outlier detection method to obtain outliers, and perform deletion, replacement, or marking processing on the outliers to obtain the current data to be corrected.

9. The online monitoring system for the current of the DC support capacitor of the flexible DC converter valve according to claim 6, characterized in that, The correction module is specifically used for: Obtain historical current waveform data, where the historical current waveform data includes normal current waveform data and abnormal current waveform data; Construct an initial neural network model for correction, and train the initial neural network model based on the historical current waveform data to obtain a trained neural network model; Input the current data to be corrected into the neural network model for processing to obtain the corrected current data to be analyzed.

10. The online monitoring system for the current of the DC support capacitor of the flexible DC converter valve according to claim 6, wherein, The analysis module is specifically used for: Extract the waveform feature data of the current waveform data, where the waveform feature data includes peak data, valley data, average data, and fluctuation range data; Construct an initial waveform recognition model, and train the initial waveform recognition model with the obtained historical waveform data and corresponding abnormal labels to obtain a trained waveform recognition model; Input the waveform features into the waveform recognition model to classify the waveform features to obtain the working state of the selected capacitor.

Images