A direct-current support capacitor capacitance online identification method based on multi-frequency point collaborative analysis

Abstract

This invention proposes an online identification method for the capacitance value of a DC-supported capacitor based on multi-frequency point collaborative analysis, relating to the field of DC-supported capacitor technology. The method includes: acquiring ripple voltage and ripple current signals; determining a delay step size and constructing a pulse weight sequence based on the delay step size; performing discrete convolution operations on the pulse weight sequence with the ripple voltage and ripple current signals to extract periodic signals; processing the periodic signals to extract signal amplitudes; determining the optimal parameter combination; obtaining optimized amplitudes; calculating a first capacitance value and a second capacitance value based on the amplitude relationship; and performing multi-frequency point collaborative analysis based on the first and second capacitance values ​​to determine the actual capacitance value of the DC-supported capacitor. This invention achieves high-precision extraction of signals at different frequency points through adaptive time-slot pulse weighting, synchronous demodulation, and integral filtering, and combines this with a multi-frequency point collaborative analysis mechanism to realize rapid and accurate identification of the capacitance parameters of the DC-supported capacitor.

Description

Technical Field

[0001] This invention relates to the field of DC-supported capacitor technology, and in particular to an online identification method for the capacitance value of DC-supported capacitors based on multi-frequency point collaborative analysis. Background Technology

[0002] DC support capacitors, as key energy storage components in power electronic equipment, play a crucial role in power transmission and transformation systems, performing core functions such as reactive power compensation, harmonic suppression, DC bus voltage stabilization, and high-frequency energy transfer. In high-voltage direct current (HVDC) converter stations, DC support capacitors maintain grid voltage stability through transient energy exchange. In flexible AC transmission devices such as static var compensators (SVCs) and static synchronous compensators (STATCOMs), DC support capacitors are used to filter characteristic subharmonics. In multilevel power converters, DC support capacitors provide a low-impedance energy buffer path, effectively suppressing DC voltage fluctuations caused by power surges. According to statistics from the International Council on Large Electric Systems (CIGRE), capacitor-related faults account for 32% of power transmission and transformation system failures, with repair costs approximately 5-8 times higher than preventative maintenance. Furthermore, capacitor bank capacity accounts for 15%-20% of the total equipment investment in ultra-high-voltage converter stations, and their reliability directly affects the system's operational economy and power supply stability.

[0003] Deterioration of DC support capacitor parameters poses a serious threat to the safe operation of power transmission and transformation systems. Capacitance decay is one of the common failure modes of DC support capacitors. Under long-term effects of high temperature, high current, and alternating electrical stress, the internal electrode polarization and dielectric aging of the DC support capacitor lead to a reduction in the effective capacitance area, resulting in a decrease in capacitance. Film capacitors are commonly used as DC support capacitors due to their low equivalent series resistance, high withstand voltage, large ripple current capacity, and long service life, making them particularly suitable for DC support applications in high-voltage, high-power power electronic equipment. When the capacitance of a DC support capacitor drops below 80% of its nominal value, a warning should be issued; if it drops below 70%, it should be replaced immediately. Abnormal capacitance not only affects the dynamic performance of equipment but also triggers a series of chain reactions: in DC transmission systems, insufficient capacitance of the DC support capacitor can cause increased fluctuations in DC bus voltage, leading to increased harmonics in the converter output; in electric vehicle charging piles, it can cause excessive output voltage ripple, increasing battery thermal stress; in smart grids, it can lead to deterioration of the response characteristics of power electronic devices, reducing system stability margin, and in severe cases, even triggering system protection actions leading to shutdown.

[0004] Traditional DC-supported capacitor capacitance monitoring technologies mainly include impedance analysis, impulse response, and spectrum analysis. Impedance analysis calculates the capacitance by measuring the impedance characteristics of the DC-supported capacitor at a specific frequency, but this method is significantly affected by external electromagnetic interference. The impulse response method, based on the excitation-response principle, is simple to operate and inexpensive, but has a long dynamic response time (>800ms) and poses potential safety risks in high-voltage scenarios. Spectrum analysis supports non-contact measurement, but is sensitive to background harmonics and requires complex digital signal processing algorithms. Systematic comparative studies show that existing methods have an average online monitoring error of 4.2%-6.8% in complex electromagnetic environments, with a response delay exceeding 500ms, which cannot meet the accuracy and speed requirements of real-time monitoring of DC-supported capacitor capacitance parameters for next-generation power electronic equipment. Therefore, a high-precision, highly interference-resistant, and fast-responding online identification technology for DC-supported capacitor capacitance is urgently needed. Summary of the Invention

[0005] In view of this, the present invention proposes an online identification method for the capacitance value of DC support capacitors based on multi-frequency point collaborative analysis. Through adaptive time-slot pulse weighting, synchronous demodulation and integral filtering, high-precision extraction of signals at different frequency points is performed. Combined with the multi-frequency point collaborative analysis mechanism, the capacitance value parameters of DC support capacitors can be identified quickly and accurately.

[0006] The technical solution of this invention is implemented as follows:

[0007] This invention provides an online identification method for the capacitance value of DC-supported capacitors based on multi-frequency point collaborative analysis, comprising:

[0008] S1. Use a sensor system to collect the ripple voltage signal and ripple current signal across the DC support capacitor;

[0009] S2. Determine the delay step size based on the power grid base frequency and sampling frequency, and construct a pulse weight sequence based on the delay step size;

[0010] S3. Perform discrete convolution operation on the pulse weight sequence with the ripple voltage signal and the ripple current signal to extract the periodic signal corresponding to the power grid fundamental frequency and its harmonic components;

[0011] S4. Perform signal processing on the periodic signal, and extract the signal amplitude for the power grid base frequency and harmonic frequency points respectively;

[0012] S5. Use a parameter search algorithm to iterate through and optimize the parameters to determine the optimal parameter combination;

[0013] S6. Re-execute steps S2 to S4 using the optimal parameter combination to obtain the amplitudes of the optimized ripple voltage signal and ripple current signal at the base frequency and harmonic frequency points of the power grid.

[0014] S7. Based on the capacitor voltage-current relationship, calculate the first capacitance value according to the amplitude relationship of the optimized ripple voltage signal and ripple current signal at the grid base frequency point, and calculate the second capacitance value according to the amplitude relationship of the optimized ripple voltage signal and ripple current signal at the harmonic frequency point.

[0015] S8. Based on the first capacitance value and the second capacitance value, perform multi-frequency point collaborative analysis to determine the actual capacitance value of the DC support capacitor.

[0016] Preferably, in step S2, the formula for determining the delay step size D is:

[0017]

[0018] in Sampling frequency, This is the fundamental frequency.

[0019] Preferably, in step S2, the pulse weight sequence The construction formula is:

[0020]

[0021] in For discrete unit pulses, M is the number of pulses, D is the delay step size, and n is the discrete time index.

[0022] Preferably, step S1 specifically includes:

[0023] The ripple voltage and ripple current signals across the DC support capacitor are acquired at a fixed sampling frequency.

[0024] The data from the first N sampling points are selected for processing, where N is a preset value, to ensure that the selected data covers one or more complete cycles of the power grid base frequency.

[0025] Preferably, in step S3, the discrete convolution operation achieves selective enhancement and noise suppression of specific frequency components by performing time-domain convolution of the pulse weight sequence with the ripple voltage signal and the ripple current signal, respectively, to obtain the weighted voltage signal and current signal.

[0026] Preferably, in step S4, the signal processing includes multiplying the extracted periodic signal by the cosine function and the sine function of the corresponding frequency to obtain the in-phase component and the quadrature component, and performing discrete integral filtering, amplitude compensation and tail-segment averaging on the in-phase component and the quadrature component.

[0027] Preferably, multiplying the extracted periodic signal with the cosine and sine functions of the corresponding frequencies means: for the power grid fundamental frequency and harmonic frequency points, multiplying the convolved voltage signal with the cosine function of the corresponding frequency to obtain the voltage in-phase component, and multiplying it with the sine function of the corresponding frequency to obtain the voltage quadrature component; multiplying the convolved current signal with the cosine function of the corresponding frequency to obtain the current in-phase component, and multiplying it with the sine function of the corresponding frequency to obtain the current quadrature component.

[0028] Preferably, the discrete integral filtering process includes:

[0029] Construct an integral filter whose impulse response is of length L. lp A uniform sequence, L lp This is the length of the integration window;

[0030] The integral filter is convolved with the in-phase and quadrature components of the voltage and current, respectively, to obtain the filtered in-phase and quadrature components of the voltage and current.

[0031] The amplitude compensation includes:

[0032] The filtered in-phase and quadrature components are amplified and compensated to correct the amplitude attenuation caused by integral filtering.

[0033] The tail segment averaging process includes:

[0034] Calculate the voltage and current amplitudes at each time point, and then perform an average calculation on the tail end of the signal to obtain the final estimated values ​​of voltage and current amplitudes.

[0035] Preferably, in step S5, the search space of the parameter search algorithm includes: the number of pulses ranges from 2 to 10, the length parameter of the integral filter ranges from 1 to 5, and the tail segment average ratio parameter ranges from 2 to 10.

[0036] Preferably, the multi-frequency point collaborative analysis in step S8 includes:

[0037] When the difference between the first capacitance value and the second capacitance value is less than a preset deviation threshold, the arithmetic mean is used as the actual capacitance value of the DC support capacitor.

[0038] When the difference between the first capacitance value and the second capacitance value is greater than or equal to a preset deviation threshold, one of the capacitance values ​​is preferentially selected as the actual capacitance value of the DC support capacitor based on the signal-to-noise ratio.

[0039] The signal-to-noise ratio is determined by the signal amplitude, spectral characteristics, or statistical characteristics of historical data.

[0040] The present invention has the following advantages over the prior art:

[0041] (1) The online identification method for the capacitance value of DC support capacitor based on multi-frequency point collaborative analysis proposed in this invention, by collecting the ripple voltage and current signals at both ends of the DC support capacitor, and combining pulse weight sequence construction, discrete convolution operation, synchronous demodulation processing and multi-frequency point collaborative analysis technology, realizes high-precision online identification of the capacitance value of DC support capacitor, effectively solving the technical problems of insufficient accuracy and lag response of existing capacitance value monitoring methods in complex electromagnetic environments;

[0042] (2) This invention employs a signal processing method that combines adaptive time-slot pulse weighting with synchronous demodulation, ensuring that the effective extraction of target frequency components is unaffected by power grid frequency fluctuations and noise interference. By performing discrete convolution operations on the acquired ripple signal and the precisely constructed pulse weight sequence, and then combining orthogonal modulation decomposition and integral filtering, the accuracy of amplitude extraction is significantly improved, keeping the capacitance parameter identification error below 1%. This represents a significant improvement compared to the 4.2%-6.8% error rate of traditional spectrum analysis methods, meeting the technical requirements of high-precision monitoring in power systems.

[0043] (3) This invention simplifies the computation process based on discrete integration and pulse weighting, reducing algorithm complexity and optimizing computational resource usage. By using a parameter search algorithm to self-optimize key parameters such as the length of the integral filter and the average proportion of the tail segment, the amount of data processing is reduced while ensuring measurement accuracy. This enables the system to achieve a response time of 30-50ms on an embedded platform, meeting the technical requirements of power systems for real-time monitoring of capacitor status and effectively supporting rapid early warning and handling of power grid faults.

[0044] (4) This invention employs a multi-frequency point collaborative analysis strategy, simultaneously processing the fundamental frequency of the power grid and its harmonic frequency components to obtain capacitance values ​​at different frequency points. The final capacitance value is then determined through threshold comparison and signal-to-noise ratio analysis. This multi-frequency point data fusion processing method effectively suppresses random interference and system errors in the power system, improving the reliability and stability of capacitance value calculation. Especially in environments with high harmonic content in the power grid, compared to single-frequency point analysis methods, multi-frequency point collaborative analysis can more accurately reflect the actual state of the capacitor, providing more precise parameter basis for capacitor life assessment. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart of the method of the present invention;

[0047] Figure 2 This is a diagram illustrating the technical implementation of the present invention. Detailed Implementation

[0048] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0049] like Figure 1 As shown, this invention provides an online identification method for the capacitance value of a DC-supported capacitor based on multi-frequency point collaborative analysis, comprising:

[0050] S1. Use a sensor system to collect the ripple voltage signal and ripple current signal across the DC support capacitor;

[0051] S2. Determine the delay step size based on the power grid base frequency and sampling frequency, and construct a pulse weight sequence based on the delay step size;

[0052] S3. Perform discrete convolution operation on the pulse weight sequence with the ripple voltage signal and the ripple current signal to extract the periodic signal corresponding to the power grid fundamental frequency and its harmonic components;

[0053] S4. Perform signal processing on the periodic signal, and extract the signal amplitude for the power grid base frequency and harmonic frequency points respectively;

[0054] S5. Use a parameter search algorithm to iterate through and optimize the parameters to determine the optimal parameter combination;

[0055] S6. Re-execute steps S2 to S4 using the optimal parameter combination to obtain the amplitudes of the optimized ripple voltage signal and ripple current signal at the base frequency and harmonic frequency points of the power grid.

[0056] S7. Based on the capacitor voltage-current relationship, calculate the first capacitance value according to the amplitude relationship of the optimized ripple voltage signal and ripple current signal at the grid base frequency point, and calculate the second capacitance value according to the amplitude relationship of the optimized ripple voltage signal and ripple current signal at the harmonic frequency point.

[0057] S8. Based on the first capacitance value and the second capacitance value, perform multi-frequency point collaborative analysis to determine the actual capacitance value of the DC support capacitor.

[0058] like Figure 2As shown, this invention provides an online identification method for the capacitance parameters of DC-supported capacitors based on multi-frequency point collaborative analysis. This method acquires the ripple voltage and current signals across the DC-supported capacitor, and combines pulse weight sequence construction, discrete convolution operation, and orthogonal synchronous demodulation processing to achieve high-precision online identification of the capacitor's capacitance value. The technical implementation process of this invention mainly includes the following five steps: A. Data acquisition and preprocessing; B. Target time pulse weighted processing; C. Orthogonal synchronous demodulation and integral filtering; D. Automatic search and optimization; E. Real-time capacitance calculation. These five steps will be described in detail below.

[0059] A. Data Acquisition and Preprocessing

[0060] First, a sensor system deployed around the DC-supported capacitor is used to synchronously acquire the ripple voltage signal across the DC-supported capacitor. and the ripple current signal flowing through the capacitor The acquired analog signals are converted into digital signal sequences by a high-speed analog-to-digital converter. and Where n represents the sampling sequence number. The system uses a fixed sampling frequency. The analog signal is discretely sampled, and the first N sampling points are selected for subsequent processing to ensure that the data covers one or more complete cycles of the power grid's base frequency. In actual implementation, the sampling frequency... It is typically set between 10kHz and 50kHz to meet the requirements of the power grid base frequency. The system requires sampling of (e.g., 50Hz or 60Hz) and its harmonic components. The system performs data window truncation on the acquired digital signal sequence; the window length is typically set to... The delay step size is an integer multiple of the time limit to include the complete periodic signal. Simultaneously, the system uses a high-pass filter to remove DC components and low-frequency interference, and a low-pass filter to suppress high-frequency noise, thereby improving the signal quality for subsequent processing. Based on the grid's fundamental frequency characteristics, the delay step size D is calculated for constructing the subsequent time-slot pulse sequence; its calculation formula is as follows:

[0061]

[0062] in Sampling frequency, This is the fundamental frequency. Taking a 10kHz sampling frequency as an example, when the fundamental frequency is 50Hz, the delay step size D=200, meaning that every 200 sampling points correspond to one fundamental frequency cycle.

[0063] B. Adaptive time-slot pulse weighted filtering

[0064] This step extracts signal components synchronized with the target frequency by constructing a pulse sequence, thereby achieving effective weighting of the original signal. Discrete convolution enhances the portion of the original signal corresponding to the target period while suppressing other noise and interference. The resulting signal has a higher signal-to-noise ratio at the target frequency, providing clearer signal characteristics for subsequent orthogonal synchronous demodulation. The specific implementation steps are as follows:

[0065] B1. Construct a time-slotted pulse weighted sequence

[0066] Based on the previously calculated delay step D, a pulse weight sequence is constructed. The sequence is a set of periodically distributed unit pulses, defined as:

[0067]

[0068] in This represents the unit impulse function, when n=0. ,otherwise, M is the number of pulses, an adjustable parameter ranging from 2 to 10, indicating how many cycles of data are used for processing; m is the pulse number, ranging from 0 to M-1; n is the discrete-time index. This sequence "focuses" on the target frequency component in the time domain, enabling more effective suppression of other frequency components and noise interference during subsequent synchronous demodulation.

[0069] B2. Discrete Convolution Operation

[0070] For input signal With pulse sequence Perform discrete convolution:

[0071]

[0072] in For input signal, It is a pulse weight sequence. is the output signal after convolution, and n is the discrete-time index.

[0073] In the frequency domain, the Fourier transform of a pulse sequence is a periodically repeating function. After convolution, it is equivalent to processing the signal using a sampling window function, thereby selectively enhancing specific frequency components in the spectrum. Specifically, the target periodic component in the signal is cumulatively enhanced only when n is aligned with the pulse position; while aperiodic or other frequency components cancel each other out due to their different phases, achieving noise suppression.

[0074] Similarly, for the input signal Performing the same process, we obtain:

[0075]

[0076] C. Quadrature synchronous demodulation and integral filtering

[0077] In this step, orthogonal modulation technology is used to decompose the signal into orthogonal components, and discrete integral filtering is applied for low-pass smoothing to remove high-frequency noise. Amplitude compensation and tail-end averaging strategies are employed to ensure the stability and accuracy of the extracted signal amplitude. The processed amplitude data provides crucial information for error analysis and capacitance calculation. Overall, this step ensures the integrity and accuracy of the signal information, directly affecting the system's measurement precision. Specific implementation includes:

[0078] C1. Using cosine and sine waves orthogonal to the target signal as references, the target frequency components are mapped onto two orthogonal components, thereby achieving separation of amplitude and phase. The formula is described as follows:

[0079]

[0080]

[0081] in, For in-phase components, For orthogonal components, For the target frequency, Where n is the sampling frequency and n is the discrete-time index.

[0082] Similarly, the same processing is applied to the current signal:

[0083]

[0084]

[0085] The in-phase component of the current signal is calculated. and orthogonal components .

[0086] C2. To eliminate high-frequency interference and noise introduced during modulation, a discrete integrator filter is used. The filter's impulse response is:

[0087]

[0088] in Let be the length of the integration window, and , Here, is the length parameter of the integral filter, D is the delay step size, and n is the discrete-time index;

[0089] Perform convolution operations on the in-phase components and quadrature components separately:

[0090]

[0091]

[0092] in, This is the filtered in-phase component. These are the orthogonal components after filtering; this is equivalent to performing a moving average on the signal to smooth out high-frequency noise.

[0093] Similarly, the same processing is applied to both the in-phase and quadrature components of the current signal:

[0094]

[0095]

[0096] C3. After synchronous demodulation, due to the energy distribution problem of sine and cosine modulation (twice the signal energy is distributed across two components), it needs to be multiplied by a compensation factor of 2:

[0097]

[0098]

[0099] The amplitude estimate after modulation compensation is obtained:

[0100]

[0101]

[0102] in, , Estimating the amplitude of voltage and current signals at each time point;

[0103] C4. To reduce the impact of initial transients or boundary effects, the signal tail segment is used for averaging:

[0104]

[0105]

[0106] in, , This refers to the final estimated signal amplitude for voltage and current signals. The starting point of the tail segment is N, and the total number of sampling points is N. Parameters for controlling the averaging ratio; This represents the number of tail sample points used for averaging calculations.

[0107] Through the above processing, the voltage amplitude at the power grid's fundamental frequency point f0 can be obtained. and current amplitude and the voltage amplitude at the harmonic frequency point 2f0 and current amplitude .

[0108] D. Automatic Search and Optimization

[0109] In this step, the system optimizes the parameters for amplitude extraction of both voltage and current signals to improve the accuracy of amplitude estimation. Because voltage and current signals have different characteristics (such as signal-to-noise ratio and harmonic content), different combinations of optimization parameters may be needed for the two signals. Specific implementation includes:

[0110] D1. Search parameters include the number of pulses M and the length of the integrator filter. Tail segment average proportion parameter The traversal process calculates the amplitude at the target frequency of each channel for each set of parameters within the preset parameter space, following the steps outlined above. .

[0111] The number of pulses M ranges from 2 to 10, and the integral filter length parameter... The value range is 1 to 5, and the value range of the tail average proportion parameter is 2 to 10.

[0112] The optimization process employs a parametric grid search method, which involves traversing every parameter combination within a preset parameter space, calculating the amplitude estimate at the target frequency of each channel according to the aforementioned steps, and then comparing it with known reference values ​​to calculate the error. The process is as follows:

[0113] The minimum error is initialized to a relatively large value;

[0114] For each set of parameter combinations in the parameter space :

[0115] Using this set of parameters, process the signal according to steps B and C to obtain the amplitude estimate. ;

[0116] Calculate the relative error for each signal:

[0117]

[0118] in, For the pre-calibrated reference amplitude, The calculated amplitude estimate;

[0119] If the relative error is less than the minimum error, then update the minimum error to the current relative error, and record the current parameter combination as the optimal combination;

[0120] Output the optimal parameter combination .

[0121] For voltage and current signals, the above optimization process is performed separately to obtain their respective optimal parameter combinations. These parameters will be used in actual measurements to ensure high accuracy in amplitude extraction.

[0122] D2. In addition to the grid search method, the system can also implement an adaptive optimization mechanism. During actual operation, the system continuously adjusts parameters through error feedback, gradually improving the amplitude extraction accuracy. The error feedback mechanism is as follows:

[0123] 1. Set error threshold ;

[0124] 2. After each measurement, compare the current amplitude estimate with historical data or expected values;

[0125] 3. If the error exceeds the threshold, a local search is performed near the current parameter to find a better parameter;

[0126] 4. Parameter updates use a weighted average method:

[0127]

[0128] in, This is a smoothing factor, ranging from 0.7 to 0.9. The better parameters found for the local search.

[0129] This adaptive optimization mechanism enables the system to continuously learn and adapt to changes in signal characteristics during operation, thereby improving the system's robustness and adaptability.

[0130] E. Real-time capacitance calculation

[0131] In this step, the real-time calculation result is compared with the reference value obtained through pre-experimental calibration to calculate the error ratio. Based on the error, the processing parameters are further adjusted to achieve automatic correction, ensuring that the amplitude error of each frequency in the final output is below a preset threshold. The system calculates the capacitance value of the DC support capacitor based on the voltage and current amplitudes obtained in the preceding steps. The specific implementation process is as follows:

[0132] E1. First, according to the basic principle of capacitors, the relationship between capacitor current and voltage is as follows:

[0133]

[0134] in, The current signal flowing through the capacitor is represented by C, where C is the capacitance value of the capacitor. denoted as , where is the rate of change of the voltage across the capacitor.

[0135] E2. When the DC support capacitor operates with sinusoidal ripple across its terminals, let the voltage signal be:

[0136]

[0137] in, This represents the amplitude of the voltage ripple. The signal frequency.

[0138] Differentiating the voltage signal, we get:

[0139]

[0140] Substituting into the current-voltage relationship of the capacitor, we obtain the current expression:

[0141]

[0142] Therefore, the amplitude of the current signal With voltage signal amplitude The following relationship exists between them:

[0143]

[0144] The formula for calculating the capacitance value can be obtained by reorganizing:

[0145]

[0146] E3. In practical applications, the system uses the voltage and current amplitudes obtained in step C and optimized in step D to calculate the capacitance value at two frequency points: the power grid fundamental frequency and the harmonic frequency. Specifically, for the 50Hz and 100Hz frequency points, the amplitudes are calculated using the aforementioned synchronous demodulation and integral filtering methods, and then substituted into the capacitance value calculation formula to obtain the capacitance value at the two frequency points:

[0147] For the fundamental frequency point, i.e., the 50Hz frequency point, calculate the first capacitance value. :

[0148]

[0149] in, This is the optimized current amplitude estimate at the fundamental frequency. This is the optimized voltage amplitude estimate at the fundamental frequency.

[0150] For the harmonic frequency point, i.e., the 100Hz frequency point, calculate the second capacitance value. :

[0151]

[0152] in, This is the optimized current amplitude estimate at the harmonic frequency point. This is the optimized voltage amplitude estimate at the harmonic frequency point.

[0153] E4. Multi-frequency point collaborative analysis. Due to various interference factors in the power system (such as harmonics, transient impulses, etc.), the capacitance value calculated at a single frequency point may contain errors. This invention adopts a multi-frequency point collaborative analysis strategy, as follows:

[0154] Calculate the relative difference in capacitance between the two frequency points:

[0155]

[0156] Set a preset threshold δ, for example, 5%.

[0157] When relative differences At that time, assuming that the capacitance calculation results at both frequency points are reliable, the arithmetic mean is used as the final capacitance value:

[0158]

[0159] When relative differences In this case, it is necessary to determine which frequency point's calculation result is more reliable. The system makes this judgment by comparing the signal-to-noise ratio (SNR) of the signals at the two frequency points, and selects the capacitance value of the frequency point with the higher SNR as the final capacitance value.

[0160]

[0161] E5. Capacitance Value Evaluation. The system will compare the calculated capacitance value C with the rated capacitance value of the DC-supported capacitor. By comparison, the health status of the DC support capacitors can be evaluated.

[0162] Specifically, in one embodiment of the present invention, a health ratio can be calculated first: Then, the health status is assessed based on the health ratio. If so, then the person is in good health; if If so, then the health status is normal; if If a health status warning is issued; if If so, it needs to be replaced.

[0163] The online identification method for capacitor capacitance parameters based on multi-frequency point collaborative analysis provided by this invention has the following technical advantages:

[0164] High-precision measurement: By combining adaptive time-slot pulse weighting with synchronous demodulation, the target frequency amplitude extraction error is reduced to less than 1%, significantly improving online measurement accuracy. Compared to the 4.2%-6.8% error rate of traditional methods in complex electromagnetic environments, this invention significantly improves the accuracy of capacitance measurement, providing a more reliable data foundation for the condition assessment of DC-supported capacitors.

[0165] Rapid Real-Time Response: Based on a simplified computational process using discrete integration and pulse weighting, the system enables real-time calculations on an embedded platform, meeting the high responsiveness requirements of online monitoring. Compared to the response delay of over 500ms in traditional methods, this invention reduces the response time to 30-50ms, significantly improving the system's real-time monitoring capability for changes in the state of DC support capacitors and effectively supporting rapid early warning and handling of power grid faults.

[0166] Multi-frequency point comprehensive correction: Simultaneous processing of fundamental and harmonic frequency components, along with multi-point data calculations, reduces the impact of external interference, resulting in more stable capacitance value calculations. Especially in power grid environments with high harmonic content, multi-frequency point collaborative analysis effectively suppresses errors that may arise from single-frequency point calculations, improving the stability and reliability of capacitance value identification and providing more comprehensive parameter data for assessing the health status of DC support capacitors.

[0167] 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 protection scope of the present invention.

Claims

1. A method for online identification of the capacitance value of a DC support capacitor based on multi-frequency point collaborative analysis, characterized in that, include: S1. Use a sensor system to collect the ripple voltage signal and ripple current signal across the DC support capacitor; S2. Determine the delay step size based on the power grid base frequency and sampling frequency, and construct a pulse weight sequence based on the delay step size; S3. Perform discrete convolution operation on the pulse weight sequence with the ripple voltage signal and the ripple current signal to extract the periodic signal corresponding to the power grid fundamental frequency and its harmonic components; S4. Perform signal processing on the periodic signal, and extract the signal amplitude for the power grid base frequency and harmonic frequency points respectively; S5. Use a parameter search algorithm to iterate through and optimize the parameters to determine the optimal parameter combination; S6. Re-execute steps S2 to S4 using the optimal parameter combination to obtain the amplitudes of the optimized ripple voltage signal and ripple current signal at the grid base frequency and harmonic frequency. S7. Based on the capacitor voltage-current relationship, calculate the first capacitance value according to the amplitude relationship of the optimized ripple voltage signal and ripple current signal at the grid base frequency point, and calculate the second capacitance value according to the amplitude relationship of the optimized ripple voltage signal and ripple current signal at the harmonic frequency point. S8. Based on the first capacitance value and the second capacitance value, perform multi-frequency point collaborative analysis to determine the actual capacitance value of the DC support capacitor; In step S5, the search space of the parameter search algorithm includes: the number of pulses ranges from 2 to 10, the length parameter of the integral filter ranges from 1 to 5, and the tail segment average ratio parameter ranges from 2 to 10.

2. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 1, characterized in that, In step S2, the formula for determining the delay step size D is: in Sampling frequency, This is the fundamental frequency.

3. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 1, characterized in that, In step S2, the pulse weight sequence The construction formula is: in For discrete unit pulses, M is the number of pulses, D is the delay step size, and n is the discrete time index.

4. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 1, characterized in that, Step S1 specifically includes: The ripple voltage and ripple current signals across the DC support capacitor are acquired at a fixed sampling frequency. The data from the first N sampling points are selected for processing, where N is a preset value, to ensure that the selected data covers one or more complete cycles of the power grid base frequency.

5. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 1, characterized in that, In step S3, the discrete convolution operation performs time-domain convolution of the pulse weight sequence with the ripple voltage signal and the ripple current signal respectively, thereby achieving selective enhancement and noise suppression of specific frequency components and obtaining weighted voltage and current signals.

6. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 1, characterized in that, In step S4, the signal processing includes multiplying the extracted periodic signal by the cosine function and sine function of the corresponding frequency to obtain the in-phase component and the quadrature component, and performing discrete integral filtering, amplitude compensation and tail-segment averaging on the in-phase component and the quadrature component.

7. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 6, characterized in that, The step of multiplying the extracted periodic signal with the cosine and sine functions of the corresponding frequencies means: for the power grid fundamental frequency and harmonic frequency points, multiplying the convolved voltage signal with the cosine function of the corresponding frequency to obtain the voltage in-phase component, and multiplying it with the sine function of the corresponding frequency to obtain the voltage quadrature component; multiplying the convolved current signal with the cosine function of the corresponding frequency to obtain the current in-phase component, and multiplying it with the sine function of the corresponding frequency to obtain the current quadrature component.

8. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 7, characterized in that, The discrete integral filtering process includes: An integrator filter is constructed with an impulse response that is a uniform sequence of length L lp , L lp is the length of the integration window. The integral filter is convolved with the in-phase and quadrature components of the voltage and current, respectively, to obtain the filtered in-phase and quadrature components of the voltage and current. The amplitude compensation includes: The filtered in-phase and quadrature components are amplified and compensated to correct the amplitude attenuation caused by integral filtering. The tail segment averaging process includes: Calculate the voltage and current amplitudes at each time point, and then perform an average calculation on the tail end of the signal to obtain the final estimated values ​​of voltage and current amplitudes.

9. The method for online identification of DC support capacitor capacitance based on multi-frequency point collaborative analysis according to claim 1, characterized in that, The multi-frequency point collaborative analysis in step S8 includes: When the difference between the first capacitance value and the second capacitance value is less than a preset deviation threshold, the arithmetic mean is used as the actual capacitance value of the DC support capacitor. When the difference between the first capacitance value and the second capacitance value is greater than or equal to a preset deviation threshold, the signal-to-noise ratio of the signals corresponding to the base frequency and harmonic frequency points of the power grid is calculated respectively, and the capacitance value corresponding to the frequency point with the higher signal-to-noise ratio is selected as the actual capacitance value of the DC support capacitor. The signal-to-noise ratio is determined by the signal amplitude, spectral characteristics, or statistical characteristics of historical data.

Images