Current transformer dielectric loss anti-interference measurement method based on variable frequency power supply

Abstract

The invention provides a current transformer dielectric loss anti-interference measurement method based on a variable frequency power supply, and relates to the technical field of power detection, and the method comprises the steps: constructing an excitation channel, collecting a background response signal, recognizing a main interference frequency through combining a frequency tracking algorithm, and dynamically constructing a frequency deviation parameter; generating a pair of non-power-frequency excitation frequency control variable-frequency power supplies to output sine wave excitation signals; excitation voltage and response current signals are synchronously collected to form sampling signals, notch frequency is set according to frequency deviation parameters, main interference is filtered out, and first processing signals are obtained; performing weighting processing by constructing a window function matched with the frequency band characteristic to generate a second processing signal; the complex amplitude and phase information at the target frequency is extracted, the complex impedance is calculated based on the complex current method, the dielectric loss value of the current transformer is finally output, and the anti-interference capability and precision of measurement are effectively improved. According to the invention, the interference in the dielectric loss measurement of the current transformer can be processed, so that the measurement precision is improved.

Description

Technical Field

[0001] This application relates to the technical field of power detection, specifically to a method for measuring the dielectric loss and anti-interference of current transformers based on variable frequency power supplies. Background Technology

[0002] The dielectric loss measurement of current transformers in variable frequency power supplies involves applying a wideband sinusoidal excitation to the current transformer using an adjustable-frequency excitation signal from the variable frequency power supply. By synchronously acquiring the amplitude and phase information of voltage and current, complex voltage and current models are constructed. Then, the equivalent impedance parameters and dielectric loss factor of the current transformer are extracted based on the complex current method. In this process, by introducing multi-frequency components into the excitation signal, the frequency domain response resolution of the current transformer insulation medium under different electrical stresses is enhanced. This effectively suppresses power frequency background interference and system harmonic noise, improving the accuracy and stability of dielectric loss measurement. This method is suitable for quantitatively assessing the aging trend and insulation reliability of current transformers under field operating conditions and is a crucial fundamental technical means for realizing condition-based maintenance and life prediction of high-voltage equipment.

[0003] When using variable frequency power supply control algorithms to measure the dielectric loss of current transformers, a differential frequency filter structure is often constructed using fixed frequency points such as 45Hz and 55Hz to suppress power frequency interference. However, in actual field situations where the power frequency interference frequency shifts slightly, for example, from 50Hz to 49.8Hz or 50.2Hz, or where the interference spectrum contains high-order harmonic components such as 150Hz and 250Hz, the filter window function designed at the fixed frequency point lacks the ability to adapt to frequency drift and spectral distortion, easily leading to the inability to effectively suppress the filter residual. In scenarios where the interference signal has non-sinusoidal components or non-stationary characteristics, this defect will further cause spectral leakage and energy diffusion in the response band, causing the target signal and the interference signal to overlap in the frequency domain, thereby introducing estimation deviations in complex amplitude and phase, affecting the accuracy and stability of the dielectric loss factor, interfering with the identification of weak dielectric loss characteristics, and in severe cases, masking the evolution signal of insulation defects, weakening the engineering applicability of the measurement algorithm in high-sensitivity condition detection.

[0004] Therefore, a method is needed to process interference in the dielectric loss measurement of current transformers, thereby improving the measurement accuracy. Summary of the Invention

[0005] This application provides a method for measuring the dielectric loss of a current transformer based on a variable frequency power supply, which can process interference in the measurement of dielectric loss of the current transformer, thereby improving the measurement accuracy.

[0006] The first aspect of this application provides a method for measuring the dielectric loss immunity of a current transformer based on a frequency converter, the method comprising: Obtain the excitation path between the primary winding of the current transformer and ground, and dynamically identify the main interference frequency in the measurement field based on the frequency tracking algorithm to construct the frequency offset parameter; A pair of non-power frequency excitation frequencies are dynamically generated based on the frequency offset parameters, and the frequency converter is controlled to output a sinusoidal excitation signal based on the non-power frequency excitation frequencies. After the sinusoidal excitation signal is input into the current transformer measurement circuit, the excitation voltage signal and the response current signal are sampled simultaneously to obtain the sampled signal. The sampling signal is adjusted based on the frequency offset parameter to remove the main interference component, thereby obtaining the first processed signal; Based on the non-power frequency excitation frequency and bandwidth characteristics, a window function is dynamically constructed and the first processed signal is weighted and processed to obtain the second processed signal; The second processed signal is subjected to frequency domain transformation, and the complex amplitude and phase information at the non-power frequency excitation frequency is extracted; Based on the complex amplitude and phase information, the dielectric loss factor of the current transformer is calculated using the complex current method, and the dielectric loss value of the current transformer is output.

[0007] Based on the above technical solutions, preferably, the step of obtaining the excitation path between the primary winding of the current transformer and ground, and dynamically identifying the main interference frequency in the measurement field based on a frequency tracking algorithm to construct frequency offset parameters, specifically includes: An excitation signal is applied between the primary winding of the current transformer and ground to establish the excitation channel; The background response signal in the excitation channel is acquired, and the background response signal includes the original sampled data of voltage signal and current signal; Spectrum extraction is performed on the background response signal to identify the frequency component with the largest amplitude in the background response signal and determine it as the main interference frequency; The frequency offset parameter is generated by performing a difference calculation between the main interference frequency and the preset ideal power frequency.

[0008] Based on the above technical solutions, preferably, the step of dynamically generating a pair of non-power frequency excitation frequencies according to the frequency offset parameters, and controlling the output of a sinusoidal excitation signal from the frequency converter according to the non-power frequency excitation frequencies, specifically includes: Based on the frequency offset parameter, among multiple non-power frequency excitation frequencies in the preset frequency candidate set, a pair of target non-power frequency excitation frequencies with the smallest overlap with the main interference frequency and harmonic frequency band are selected, and the frequency difference between the multiple non-power frequency excitation frequencies satisfies the frequency domain decoupling condition and the system frequency resolution constraint. The target non-power frequency excitation frequency is input into the waveform modulation module to control the frequency converter to call the waveform parameters corresponding to the target non-power frequency excitation frequency; Based on the waveform parameters, the waveform modulation module generates a pair of sinusoidal excitation signals with stable frequency, consistent amplitude, and synchronized phase.

[0009] Based on the above technical solution, preferably, after inputting the sinusoidal excitation signal into the current transformer measurement circuit, the excitation voltage signal and the response current signal are sampled simultaneously to obtain the sampled signal, specifically including: The sinusoidal excitation signal is input to the excitation channel formed between the primary winding of the current transformer and ground through the isolation drive unit; The response current signal is obtained by shorting and grounding the secondary winding terminals of the current transformer to form a complete closed measurement circuit, and then using a current detection device. By arranging voltage sampling channels at both ends of the excitation channel, the excitation voltage signal is acquired through a voltage sensor. The excitation voltage signal and the response current signal are respectively subjected to parallel analog-to-digital conversion. The excitation voltage signal and the response current signal after analog-to-digital conversion are constructed into a data structure aligned with the time sequence to form the sampling signal.

[0010] Based on the above technical solutions, preferably, the step of dynamically constructing a window function based on the non-power frequency excitation frequency and frequency band characteristics and weighting the first processed signal to obtain the second processed signal specifically includes: The first processed signal is used to construct a multidimensional cost function based on a family of cosine lifting window functions. The multidimensional cost function aims to maximize the spectral purity of the excitation frequency and minimize the energy leakage of adjacent frequency bands. According to the non-power frequency excitation frequency output frequency band characteristic parameters, the frequency band characteristic parameters include the frequency density and main interference harmonic distribution on both sides of the non-power frequency excitation frequency; A multi-objective parameter optimization algorithm is executed to generate an optimal window function sample sequence based on the non-power frequency excitation frequency and the frequency band characteristic parameters; The window function sample sequence is input into the multidimensional cost function, and a time-domain point-to-point weighting operation is performed on the first processed signal to output the second processed signal.

[0011] Based on the above technical solutions, preferably, the step of inputting the window function sample sequence into the multidimensional cost function, performing a time-domain point-to-point weighting operation on the first processed signal, and outputting the second processed signal specifically includes: Perform point-by-point multiplication on the weight coefficients at corresponding positions in the window function sample sequence for each sampling point in the first processed signal to construct a weighted result sequence; During the point-by-point product process, the frequency density gradient mapping function is called to perform dynamic edge correction on the weighted result sequence. The frequency density gradient mapping function is generated based on the energy distribution trend of the excitation frequency in the frequency band characteristic parameters. The weighted result sequence after dynamic edge correction is used as the second processing signal.

[0012] Based on the above technical solutions, preferably, the step of calculating the dielectric loss factor of the current transformer based on the complex amplitude and phase information using the complex current method, and outputting the dielectric loss value of the current transformer, specifically includes: The complex amplitude is determined to include the complex amplitude of the excitation voltage and the complex amplitude of the response current, wherein the complex amplitude of the excitation voltage includes a first amplitude and a first phase, and the complex amplitude of the response current includes a second amplitude and a second phase; The complex impedance value is calculated based on the complex amplitude of the excitation voltage and the complex amplitude of the response current. The complex impedance value is obtained by dividing the complex value of the excitation voltage formed by the first amplitude and the first phase by the complex value of the response current formed by the second amplitude and the second phase. The complex impedance value is decomposed into a real component and an imaginary component, where the real component corresponds to the resistive component and the imaginary component corresponds to the capacitive reactance component. The ratio of the resistive component to the capacitive reactance component is used as the dielectric loss factor of the current transformer. The dielectric loss factor of the current transformer is output as the dielectric loss value of the current transformer.

[0013] A second aspect of this application provides a current transformer dielectric loss immunity measurement device based on a variable frequency power supply. The device is used to perform a current transformer dielectric loss immunity measurement method based on a variable frequency power supply as described in any of the above-described methods. The device includes an acquisition module, a processing module, and an output module, wherein: The acquisition module is used to acquire the excitation path between the primary winding of the current transformer and ground, and dynamically identify the main interference frequency in the measurement site based on the frequency tracking algorithm to construct frequency offset parameters. The processing module is used to dynamically generate a pair of non-power frequency excitation frequencies according to the frequency offset parameters, and control the frequency converter to output a sine wave excitation signal according to the non-power frequency excitation frequencies. The processing module is used to input the sinusoidal excitation signal into the current transformer measurement circuit, and then simultaneously sample the excitation voltage signal and the response current signal to obtain the sampled signal. The processing module is used to adjust the notch frequency of the sampled signal based on the frequency offset parameter to filter out the main interference component, thereby obtaining a first processed signal; The processing module is used to dynamically construct a window function based on the non-power frequency excitation frequency and frequency band characteristics, and to weight and process the first processed signal to obtain the second processed signal. The processing module is used to perform frequency domain transformation on the second processed signal and extract the complex amplitude and phase information at the non-power frequency excitation frequency; The output module is used to calculate the dielectric loss factor of the current transformer based on the complex amplitude and the phase information using the complex current method, and output the dielectric loss value of the current transformer.

[0014] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any of the foregoing.

[0015] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed, perform the method described in any of the preceding descriptions.

[0016] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. This application introduces a frequency tracking algorithm to dynamically identify the main interference frequency and construct frequency offset parameters, enabling the variable frequency power supply to output non-power frequency excitation signals in a frequency band that avoids the main interference frequency and its harmonics. It also combines a notch filter to adaptively suppress the main interference frequency components, uses a dynamically constructed window function to weight the sampled signal to reduce spectral leakage, and finally extracts complex amplitude and phase information through frequency domain transformation and calculates the dielectric loss factor based on the complex current method. This effectively improves the frequency domain resolution and result stability of current transformer dielectric loss measurement under complex interference background, thereby significantly improving measurement accuracy.

[0017] 2. By applying an excitation signal through the excitation channel and acquiring the background response signal, and then combining spectrum extraction and main interference frequency identification, it is possible to accurately track the interference frequency on site, dynamically construct frequency offset parameters, effectively improve the system's response capability and adaptability to changes in interference sources, and provide an accurate basis for subsequent non-power frequency excitation frequency interference avoidance configuration.

[0018] 3. By selecting a pair of non-power frequency excitation frequencies with the lowest interference overlap from the frequency candidate set based on the frequency offset parameter, and controlling the output of excitation signals with consistent amplitude and synchronized phase from the frequency converter, the focusing of the excitation signal on the target frequency band is enhanced, the risk of spectral domain coupling between the excitation frequency and the main interference frequency is significantly reduced, and the frequency domain purity of the measurement signal is improved.

[0019] 4. By arranging voltage and current sampling paths in the excitation channel, the excitation voltage and response current signals are acquired synchronously and a time-series aligned data structure is constructed. This ensures the phase consistency between the excitation and response quantities, guarantees the calculation accuracy of subsequent frequency domain analysis and complex parameter extraction, and effectively suppresses the estimation bias caused by time domain sampling delay.

[0020] 5. By using a family of cosine lifting window functions to construct a multidimensional cost function and combining it with frequency band characteristic parameters for window function optimization, the spectral purity at the excitation frequency and the leakage energy in adjacent frequency bands are effectively enhanced, thereby improving the extraction resolution of target frequency components in frequency domain analysis and improving the accuracy of dielectric loss calculation.

[0021] 6. When weighting the first processed signal using a window function, a frequency density gradient mapping function is introduced for edge correction. The weight distribution is dynamically adjusted according to the frequency energy distribution trend within the frequency band, which effectively suppresses the spectral distortion problem caused by the window function edge, enhances the energy concentration of the main lobe of the excitation frequency, and improves the stability of complex amplitude and phase information extraction.

[0022] 7. By extracting the complex amplitude and phase information of the excitation voltage and response current, the complex impedance is calculated and decomposed into resistive and capacitive reactance components, and finally the dielectric loss factor is output. This measurement method can accurately characterize the dielectric loss characteristics of the current transformer insulation system and is suitable for quantitative evaluation and trend diagnosis of equipment insulation performance under field interference background. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating a current transformer dielectric loss anti-interference measurement method based on a frequency converter power supply disclosed in an embodiment of this application. Figure 2 This is a schematic diagram of a current transformer dielectric loss anti-interference measurement device based on a frequency converter power supply disclosed in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application.

[0024] Explanation of reference numerals in the attached drawings: 201, acquisition module; 202, processing module; 203, output module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0026] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0027] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0028] The dielectric loss measurement of current transformers in variable frequency power supplies involves applying an adjustable frequency sinusoidal excitation and simultaneously acquiring the amplitude and phase information of voltage and current. Based on the complex current method, the equivalent impedance parameter and dielectric loss factor are extracted, which has good frequency domain resolution and anti-power frequency interference capability. However, the commonly used fixed frequency differential filtering method lacks adaptive capability under the conditions of power frequency drift and high-order harmonic interference in the field, resulting in spectral leakage and complex estimation deviation, reducing the accuracy and stability of the measurement results, and limiting the ability to identify insulation degradation trends. Therefore, it is urgent to introduce an interference suppression method that can dynamically cope with frequency disturbances and non-stationary interference to improve the accuracy and engineering applicability of dielectric loss measurement.

[0029] This embodiment discloses a method for measuring the dielectric loss and interference immunity of a current transformer based on a frequency converter power supply, referring to... Figure 1 This includes the following steps S110-S170: S110 obtains the excitation path between the primary winding of the current transformer and ground, and dynamically identifies the main interference frequency in the measurement field based on the frequency tracking algorithm to construct the frequency offset parameter.

[0030] The current transformer dielectric loss anti-interference measurement method based on variable frequency power supply disclosed in this application is applied to a server. The server includes, but is not limited to, electronic devices such as mobile phones, tablets, wearable devices, and PCs (personal computers), and can also be a backend server running a current transformer dielectric loss anti-interference measurement method based on variable frequency power supply. The server can be implemented using a standalone server or a server cluster composed of multiple servers.

[0031] In one possible implementation, the excitation channel between the primary winding of the current transformer and ground is obtained, and the main interference frequency in the measurement field is dynamically identified based on a frequency tracking algorithm to construct frequency offset parameters. Specifically, this includes: applying an excitation signal between the primary winding of the current transformer and ground to construct an excitation channel; acquiring the background response signal in the excitation channel, the background response signal including the original sampled data of voltage and current signals; performing spectrum extraction on the background response signal to identify the frequency component with the largest amplitude in the background response signal and determine the main interference frequency; and performing a difference calculation between the main interference frequency and a preset ideal power frequency to generate frequency offset parameters.

[0032] Specifically, a set of sinusoidal excitation signals with frequency stability and amplitude adjustment capability is output by the frequency converter power supply control module. This signal is then connected to the lead terminals between the primary winding of the current transformer and ground via an isolated drive channel, forming an excitation channel. In this embodiment, the excitation channel refers to the electrical closed path formed by the primary winding of the current transformer, with one end being the excitation signal source output and the other end being the ground terminal. To ensure controllable impedance characteristics of the excitation channel, an isolation transformer or low-impedance buffer network is typically installed between the frequency converter power supply output and the current transformer to stabilize the signal amplitude and suppress reflection interference.

[0033] After the excitation signal is applied, the synchronous sampling module synchronously samples the voltage at both ends of the excitation channel and the response current in the series channel. The synchronous sampling module consists of a dual-channel high-precision analog-to-digital converter and uses unified clock control to achieve time-aligned sampling of the excitation voltage and response current. The background response signal refers to the voltage and current signals collected before the formal measurement excitation is applied or under low-amplitude excitation, used to analyze the existing interference spectrum distribution in the field. The raw sampling data refers to the voltage and current numerical sequence output by the sampling module without any filtering, transformation, or weighting processing, retaining all spectral information.

[0034] The acquired background response signal is input to the frequency tracking module, which contains an improved sliding Fourier transform unit. This unit performs discrete Fourier transform analysis on the signal within multiple short-time sliding windows, extracting the frequency amplitude spectrum in different time segments and statistically analyzing the frequency positions corresponding to the amplitude peaks. In identifying the frequency component with the largest amplitude, the maximum power density criterion is used; that is, the maximum value point corresponding to the frequency coordinate in the power spectral density function is found. This frequency is defined as the dominant interference frequency in the current measurement environment. The dominant interference frequency typically originates from power grid frequency, switch resonance, or high-frequency harmonic coupling components in the field, and is the frequency component with the highest interference risk in identifying the excitation signal.

[0035] The difference between the main interference frequency value output by the frequency tracking module and the ideal power frequency value set in the system (usually 50Hz) is calculated to obtain the main interference frequency offset, called the frequency offset parameter. The frequency offset parameter measures the deviation of the actual interference frequency from the ideal power frequency and serves as an input parameter for the subsequent excitation frequency avoidance strategy of the frequency converter control module. The frequency offset parameter is usually a real number, which can be positive or negative, representing the upward or downward offset trend of the interference frequency relative to the power frequency. For example, if the measured main interference frequency is 50.2Hz and the ideal power frequency is 50Hz, the frequency offset parameter is 0.2Hz. In subsequent frequency scheduling, this parameter will guide the non-power frequency excitation frequency to avoid the frequency range between 49.8Hz and 50.2Hz, thereby achieving frequency domain isolation between the excitation signal and the main interference frequency band. This parameter is the unified reference for adaptive excitation frequency control and notch filter tuning.

[0036] S120 dynamically generates a pair of non-power frequency excitation frequencies based on the frequency offset parameters, and controls the output of the frequency converter power supply to produce a sine wave excitation signal based on the non-power frequency excitation frequencies.

[0037] In one possible implementation, a pair of non-power frequency excitation frequencies are dynamically generated based on a frequency offset parameter, and the frequency converter outputs a sinusoidal excitation signal based on the non-power frequency excitation frequencies. Specifically, this includes: selecting a pair of target non-power frequency excitation frequencies with the smallest overlap with the main interference frequency and harmonic frequency band from multiple non-power frequency excitation frequencies in a preset frequency candidate set based on the frequency offset parameter, wherein the frequency difference between the multiple non-power frequency excitation frequencies satisfies the frequency domain decoupling condition and the system frequency resolution constraint; inputting the target non-power frequency excitation frequencies into a waveform modulation module, controlling the frequency converter to call the waveform parameters corresponding to the target non-power frequency excitation frequencies; and generating a pair of sinusoidal excitation signals with stable frequency, consistent amplitude, and synchronized phase based on the waveform parameters by the waveform modulation module.

[0038] Specifically, the system first establishes a candidate set containing multiple selectable non-power frequency excitation frequencies. This candidate set is set by the frequency scheduling module and typically covers multiple frequency nodes in increments of 0.1 Hz within the 40 Hz to 70 Hz range. After receiving the frequency offset parameters, the frequency scheduling module compares the overlap between each candidate frequency and the main interference frequency and its integer multiples of harmonic frequencies (e.g., 150 Hz, 250 Hz) in the spectral space. It then uses the minimum spectral overlap criterion to assess the interference risk of each candidate frequency pair and selects the target non-power frequency excitation frequency pair with the smallest frequency overlap area using frequency domain energy separation as the evaluation index. The frequency difference must satisfy the frequency domain decoupling condition of the system, meaning the interval between the two excitation frequencies must be greater than the system's minimum frequency resolution, generally determined by the sampling rate and window length. For example, with a sampling rate of 10 kHz and a window length of 2048 points, the system frequency resolution is approximately 4.88 Hz; therefore, the excitation frequency interval must be at least 5 Hz to ensure that aliasing does not occur during the extraction of independent components in the frequency domain.

[0039] The target non-power frequency excitation frequency is transmitted to the waveform modulation module via the excitation frequency scheduling interface. As the digital control core of the variable frequency power supply, the waveform modulation module stores a set of waveform synthesis parameters matched to different frequencies, including frequency register values, amplitude coefficients, initial phase angles, and sampling step sizes. Based on the received frequency index, the waveform modulation module extracts the waveform parameters for the corresponding frequency point from its built-in lookup table and uses these parameters to drive a numerically controlled oscillator (NCO) or a direct digital synthesizer (DDS) to generate a sinusoidal signal waveform in real time. This control process is implemented through programmable logic, allowing for rapid switching between multiple frequencies without interrupting the output, effectively meeting the requirements of dynamic frequency avoidance and signal continuity.

[0040] Based on the aforementioned frequency scheduling and waveform parameters, the waveform modulation module uses a dual-channel NCO to generate two sinusoidal signals, corresponding to the low-frequency and high-frequency ends of the target non-power frequency excitation frequency, respectively. The amplitude of the two signals is calibrated by an amplitude controller to ensure consistency within the error range, with amplitude deviation controlled to within ±0.2%. Simultaneously, a phase-locking mechanism unifies the initial phase of the two excitation signals, and a real-time monitoring mechanism maintains phase synchronization, preventing phase drift due to clock offset or module response differences. The two output sinusoidal excitation signals are used for the voltage excitation channel and phase reference channel, respectively, providing a highly stable excitation reference basis in subsequent complex amplitude and phase calculations, ensuring the consistency and repeatability of complex parameters in dielectric loss factor extraction.

[0041] S130: After the sinusoidal excitation signal is input into the current transformer measurement circuit, the excitation voltage signal and the response current signal are sampled simultaneously to obtain the sampled signal.

[0042] In one possible implementation, after inputting the sinusoidal excitation signal into the current transformer measurement circuit, the excitation voltage signal and the response current signal are sampled simultaneously to obtain the sampled signal, specifically including: A sinusoidal excitation signal is input to the excitation channel formed between the primary winding of the current transformer and ground through an isolation drive unit. A complete closed measurement circuit is formed by shorting and grounding the terminals of the secondary winding of the current transformer, and the response current signal is obtained through a current detection device. Voltage sampling channels are arranged at both ends of the excitation channel, and the excitation voltage signal is obtained through a voltage sensor. The excitation voltage signal and the response current signal are converted into digital signals in parallel. The converted excitation voltage signal and the response current signal are constructed into a data structure aligned with the time sequence to form the sampling signal.

[0043] Specifically, the sinusoidal excitation signal is input to the excitation channel formed between the primary winding of the current transformer and ground through an isolation drive unit. The excitation channel refers to the closed conductive path formed between the primary winding of the current transformer and ground, used to carry the injection process of the external frequency converter excitation signal. During this process, to prevent signal distortion or damage to the measurement system caused by high-voltage coupling, common-mode interference, or ground potential fluctuations, the sinusoidal excitation signal output from the frequency converter power supply needs to be isolated and buffered through the isolation drive unit. The isolation drive unit typically consists of an isolation transformer, a differential-mode amplifier, and a low-pass protection network, providing not only electrical isolation but also ensuring stable amplitude and undistorted waveform during the drive process. In practical implementation, the excitation signal, after being synthesized by a digitally controlled oscillator, is first sent to the isolation drive channel via an amplitude modulator, and finally connected to the high-voltage side of the primary winding of the current transformer through a shielded wire, with the other end grounded, forming a complete excitation path.

[0044] A complete closed measurement circuit is formed by shorting and grounding the secondary winding terminals of the current transformer, and the response current signal is acquired through a current detection device. To ensure a stable magnetic field response within the current transformer caused by the excitation signal, all secondary winding terminals must be shorted and grounded to create a closed magnetic flux path for the excitation current within the internal core. This structure simultaneously stabilizes the load and eliminates induced potential drift. The response current signal refers to the alternating current generated by the excitation voltage and flowing through the primary winding. This current reflects the capacitive, resistive, and dielectric response characteristics of the current transformer's insulation system. By connecting a current detection device, such as a high-bandwidth, low-impedance current transformer or a precision current sampling resistor, in series in the primary winding circuit and connecting its output to a sampling module, the response current signal can be acquired in real time for subsequent calculations of complex impedance and dielectric loss.

[0045] The excitation voltage signal is acquired by arranging voltage sampling channels at both ends of the excitation channel and using a voltage sensor. To accurately measure the excitation voltage applied to the primary winding of the current transformer, a voltage sampling channel needs to be established between the excitation signal injection point and ground. This voltage sampling channel includes a high-voltage divider network, a voltage buffer, and a low-pass filter, used to convert the excitation signal into a low-amplitude equivalent signal acceptable to the sampling system and to eliminate high-frequency spike interference. In high-voltage testing scenarios, a combination of a high-impedance precision voltage probe and an isolation amplifier is often used to extract the voltage across the primary winding and send it to the analog sampling front end, ensuring that the amplitude and phase characteristics of the excitation voltage signal are not corrupted or delayed before entering the analog-to-digital conversion.

[0046] The excitation voltage signal and response current signal are converted to digital signals in parallel. The converted signals are then structured into a time-series aligned data structure to form a sampled signal. Both signals are fed into a dual-channel analog-to-digital converter (ADC). This ADC uses a unified sampling clock to ensure that both channels complete the analog-to-digital conversion at the same sampling time, preventing phase mismatch or time drift. The converted digital signal is organized into a two-dimensional data structure with a fixed time step and channel labels. Each sampling point contains a pair of voltage and current values, forming a structured sampled signal. This sampled signal serves as the foundational input data source for subsequent frequency identification, filtering, and complex current method calculations. It maintains a precise correspondence in both time and channel dimensions, ensuring the accuracy of dielectric loss calculation. In a typical embodiment, a 24-bit high-resolution ADC with a sampling frequency of 10kHz is used to synchronously sample the voltage and current signals 10,000 times per second. The sampled data is sent to the signal processing module in floating-point format for further processing.

[0047] S140, the sampling signal is adjusted based on the frequency offset parameter to remove the main interference component and obtain the first processed signal.

[0048] In one possible implementation, the sampling signal is adjusted based on a frequency offset parameter to filter out the main interference component, resulting in a first processed signal. Specifically, this includes: inputting the sampling signal structure to a notch filter control module; transmitting the frequency offset parameter of the main interference frequency output by a frequency identification module to the notch filter control module; calculating the center frequency of the current main interference component based on the frequency offset parameter, and setting the center frequency and bandwidth parameters of the notch filter; inputting the notch filter parameters to an adaptive notch filter coefficient generation unit, which uses a normalized minimum mean square error algorithm to adjust the notch filter weight coefficients in real time, constructing the notch filter impulse response function; inputting the sampling signal structure to a notch filter execution unit, which then calls the notch filter impulse response function to perform time-domain convolution processing on the sampling signal structure; and outputting the first processed signal after notch filtering. The first processed signal has minimum spectral energy at the main interference frequency and retains the amplitude and phase characteristics of the excitation frequency component and the response frequency component.

[0049] Specifically, the sampled signal structure is input to the notch filter control module. The sampled signal structure is a time-aligned two-dimensional digital data sequence generated by the synchronous sampling module, where each record contains the excitation voltage signal and response current signal acquired at the same sampling time. This structure is input to the notch filter control module as complete raw measurement data. The notch filter control module is an adaptive signal processing control unit whose function is to construct a notch filter parameter configuration that can dynamically tune the center frequency and drive subsequent filtering operations.

[0050] The frequency offset parameter of the main interference frequency output by the frequency identification module is transmitted to the notch filter control module. The frequency identification module performs spectral analysis on the background response signal to extract the main interference frequency corresponding to the maximum energy density and calculates the difference between this main interference frequency and the ideal power frequency, forming the frequency offset parameter. For example, if the detected frequency is 50.2Hz and the ideal power frequency is 50Hz, the frequency offset parameter is 0.2Hz. This frequency offset parameter is transmitted in real time to the notch filter control module to determine the center position of the interference frequency.

[0051] The notch filter control module calculates the center frequency of the current main interference component based on the frequency offset parameter and sets the center frequency and bandwidth parameters of the notch filter. The notch filter control module then superimposes the frequency offset parameter with the power frequency reference value to obtain the center frequency of the current main interference frequency, denoted as . ,in For ideal power frequency, This is the frequency offset parameter. The bandwidth parameter is determined based on the stability of the main interference frequency and the power spectrum spread. Typically, a notch filter range of approximately ±0.3Hz to ±1Hz above and below the center frequency is used to ensure effective coverage of the interference frequency band and avoid interference excitation frequency neighborhood components.

[0052] The notch filter parameters are input to the adaptive notch filter coefficient generation unit, which uses a normalized minimum mean square error algorithm to adjust the notch filter weight coefficients in real time, thus constructing the notch filter impulse response function. This adaptive algorithm is applied to time series samples. With expected output Based on this, minimize the error ,in For the filtered output, the weight coefficients are updated using the normalized minimum mean square error algorithm according to the following recursive formula:

[0053] in, For the first The filter weight vector of the next iteration Step size factor To avoid positive decimals with a denominator of zero, this algorithm can adapt to slight frequency variations and converge quickly, forming the impulse response function. It is used for notch filtering operations.

[0054] The sampled signal structure is input to the notch filter execution unit, which then calls the notch filter impulse response function to perform time-domain convolution processing on the sampled signal structure. The notch filter execution unit utilizes the constructed impulse response function... For the input sampled signal structure Perform discrete convolution:

[0055] in The length of the notch filter. This is the filtered output signal. During this process, the output signal at each moment is the sum of the current input data point and the previous data point. The sample points are obtained by weighted summation of impulse responses, and interference components located in the notch band are filtered out.

[0056] The output is a first processed signal after notch filtering. This first processed signal has minimal spectral energy at the main interference frequency and retains the amplitude and phase characteristics of the excitation and response frequency components. After notch filtering, the signal energy in the interference band (e.g., 50.2Hz ± 0.5Hz) is greatly suppressed, with a spectral energy reduction of over 40dB. However, the spectral components at the excitation frequency pair (e.g., 45Hz and 55Hz) are not weakened, maintaining the original amplitude and phase structure. This ensures the reliability and accuracy of the input signal in subsequent frequency domain transformations and complex current method calculations. The output of this step serves as the first processed signal and enters the subsequent window function weighting and frequency domain decoupling process.

[0057] S150: Based on the non-power frequency excitation frequency and bandwidth characteristics, a window function is dynamically constructed and the first processed signal is weighted and processed to obtain the second processed signal.

[0058] In one possible implementation, a window function is dynamically constructed based on the non-power frequency excitation frequency and frequency band characteristics, and a first processed signal is weighted and processed to obtain a second processed signal. Specifically, this includes: constructing a multidimensional cost function based on a family of cosine lifting window functions for the first processed signal, wherein the multidimensional cost function aims to maximize the spectral purity of the excitation frequency and minimize energy leakage in adjacent frequency bands; outputting frequency band characteristic parameters based on the non-power frequency excitation frequency, wherein the frequency band characteristic parameters include the frequency density and main interference harmonic distribution on both sides of the non-power frequency excitation frequency; executing a multi-objective parameter optimization algorithm to generate an optimal window function sample sequence based on the non-power frequency excitation frequency and frequency band characteristic parameters; inputting the window function sample sequence into the multidimensional cost function, performing a time-domain point-to-point weighting operation on the first processed signal, and outputting the second processed signal.

[0059] In one possible implementation, the window function sample sequence is input into a multidimensional cost function, and a time-domain point-to-point weighting operation is performed on the first processed signal to output a second processed signal. Specifically, this includes: performing a point-by-point product operation on the weight coefficients at corresponding positions in the window function sample sequence for each sampling point in the first processed signal to construct a weighted result sequence; during the point-by-point product process, a frequency density gradient mapping function is called to perform dynamic edge correction on the weighted result sequence, the frequency density gradient mapping function being generated based on the energy distribution trend of frequencies adjacent to the excitation frequency in the frequency band characteristic parameters; and using the weighted result sequence after dynamic edge correction as the second processed signal.

[0060] Specifically, a point-by-point multiplication operation is performed on the weight coefficients at corresponding positions in the window function sample sequence for each sampling point in the first processed signal to construct a weighted result sequence. In this step, the first processed signal is a one-dimensional time series. The window function sample sequence is a weighted sequence of the same length. At every point in time, both Perform point-to-point multiplication on the above, and output a weighted result sequence. The purpose of introducing the window function is to apply non-uniform weighting to the signal in the time domain to control frequency domain leakage, thereby improving spectral resolution and suppressing interference from non-target frequency components on the estimation of the excitation frequency spectrum. The window function sample sequence is derived from the previous optimization result and has customized edge attenuation characteristics, minimizing the impact of the signal ends on the spectral structure.

[0061] During point-by-point multiplication, the frequency density gradient mapping function is invoked to dynamically correct the edges of the weighted result sequence. This function is generated based on the energy distribution trend of frequencies near the excitation frequency in the frequency band characteristic parameters. The frequency density gradient mapping function is a dynamic adjustment function used to enhance the edge selectivity of the weighting window. Essentially, it constructs an edge suppression factor based on the frequency energy distribution gradient on both sides of the excitation frequency. This allows for selective enhancement or reduction of the weights at the window function edges when the energy distribution in the vicinity of the non-power frequency excitation frequency is abnormal, preventing spectral leakage from concentrating in the interference bandwidth. The construction process includes: performing a spectral scan within the range of ±Δf from the center of the excitation frequency and calculating the power change gradient within a unit frequency bandwidth.

[0062] Then mapped to the time-domain window function edge control factor The weighted sequence is This edge correction not only considers the shape of the time-domain window, but also reflects the feedback adjustment mechanism of frequency-domain features on the window function structure, thereby improving the window function's directivity to the target frequency and its adaptability to frequency-domain interference.

[0063] The weighted result sequence after dynamic edge correction is used as the second processed signal. This output, the second processed signal, has higher frequency domain selectivity and concentrated main lobe characteristics, serving as the basis for the input signal of subsequent frequency domain transformation. Through dynamic edge correction, the second processed signal exhibits a significantly increased main lobe energy proportion at the excitation frequency and a substantial reduction in side lobe leakage in the frequency domain, while preserving the phase integrity of the original signal. In practical applications, when the target excitation frequency is 47Hz, the main interference frequency is 49.8Hz, and the edge frequency density exhibits an exponential upward trend, the frequency density gradient mapping function can dynamically suppress the weights at both ends of the window function by up to 20%, thereby enabling the second processed signal to achieve a signal-to-noise ratio gain of 5.1dB at the 47Hz frequency point, enhancing the stability and accuracy of subsequent complex amplitude and phase extraction.

[0064] Specifically, a multidimensional cost function is constructed based on a family of cosine-lifting window functions for the first processed signal. This multidimensional cost function aims to maximize the spectral purity of the excitation frequency and minimize energy leakage in adjacent frequency bands. The family of cosine-lifting window functions is a class of weighted window functions with good main lobe focusing ability and sidelobe suppression characteristics in the frequency domain; typical examples include the Kaiser window, Tukey window, and Dolph-Chebyshev window. For the first processed signal, given that the signal energy in the frequency domain is mainly concentrated at non-power frequency excitation frequencies, a multidimensional cost function needs to be introduced into the window function design to ensure that the excitation frequency energy is not interfered with by spectral leakage while reducing the spread of signal energy to adjacent frequency bands. To achieve the optimization objectives of maximizing the spectral purity of the excitation frequency and minimizing energy leakage in adjacent frequency bands, the following normalized multidimensional cost function is defined. ,in The window function parameter vector (e.g., window width coefficient, shape parameter) is

[0065] Among them The complex spectrum in the frequency domain after windowing the first processed signal and performing a Fast Fourier Transform. This is a non-power frequency excitation frequency. For frequency analysis of total bandwidth, Leakage frequency bands adjacent to the excitation frequency (e.g.) (interval) , These are the weighting coefficients for the main frequency preservation term and the sideband suppression term in the cost function. The goal is to satisfy... Find the optimal window function parameters under the condition of minimization. The cost function comprises two objective components: first, maximizing the spectral amplitude at the excitation frequency; and second, minimizing the energy integral within a preset bandwidth on both sides of that frequency. Through this cost function constraint, the window function optimization process not only improves the clarity of the main frequency components but also suppresses energy leakage in the frequency domain sidebands, thereby reducing interference from non-target frequency bands on subsequent complex number calculations.

[0066] Based on the output band characteristic parameters of the non-power frequency excitation frequency, the band characteristic parameters include the frequency density and the distribution of the main interference harmonics on both sides of the non-power frequency excitation frequency. These band characteristic parameters describe the background spectrum distribution of the first processed signal in the frequency domain, specifically including the energy density (i.e., frequency density) per unit bandwidth in the adjacent frequency bands to the left and right of the excitation frequency point, and the peak values ​​of the interference harmonics located at integer multiples or linear combinations of the excitation frequency point. In this step, the signal analysis module performs a short-time Fourier transform on the first processed signal and scans the spectral curve within ±10Hz of the target excitation frequency, extracts the corresponding energy density curve, calculates the maximum spectral value and standard deviation within the interference frequency interval, forms the main interference harmonic distribution index, and constructs a band characteristic vector by combining the spectral gradient information of the excitation frequency bandwidth boundary, which serves as the input feature for subsequent window function parameter tuning.

[0067] A multi-objective parameter optimization algorithm is executed to generate an optimal window function sample sequence based on the non-power frequency excitation frequency and frequency band characteristic parameters. This optimization process, based on multi-objective optimization algorithms such as particle swarm optimization or gradient descent, dynamically adjusts control parameters, such as window width, edge attenuation rate, and main lobe width, within a pre-defined family of cosine lifting window functions to adapt to the current frequency band characteristic parameters. The optimization process iteratively calculates the cost function values ​​under different window functions, seeking the optimal set of window function parameters that maximizes the peak of the excitation frequency spectrum and minimizes leakage energy in adjacent frequency bands. Finally, a set of time-domain window function sample sequences is derived from the optimization results. The sample sequence length is consistent with the first processed signal, and the values ​​are typically between 0 and 1, representing the weight distribution of different time sampling points.

[0068] The window function sample sequence is input into a multidimensional cost function, and a time-domain point-to-point weighting operation is performed on the first processed signal to output the second processed signal. In this step, the window function weighting module uses the window function sample sequence as a time-domain weighting template and performs point-by-point multiplication with the first processed signal. That is, the excitation voltage value and response current value at each time step are multiplied by the corresponding window function weight to form a new weighted sample value. This operation is equivalent to weighted attenuation at both ends of the signal in the time domain, while maintaining the energy concentration of the main lobe in the middle. Mathematically, the second processed signal is equal to the product of the sample value of the first processed signal at each time step and the corresponding weight value in the window function sample sequence. This operation ensures that the signal energy at the excitation frequency is compressed and concentrated in the frequency domain, while reducing the spectral diffusion introduced by edge data points, and enhancing the amplitude accuracy and phase stability of the frequency domain extraction results. In the embodiment, when the excitation frequency is 45Hz, there is a main interference peak of 49.8Hz to 50.2Hz in the frequency band. By optimizing the generated Kaiser window function sample, the main lobe peak of 45Hz can be increased by 4.3dB, the energy of the interference frequency band can be reduced by 32.6dB, and the main frequency and interference frequency components can be effectively isolated.

[0069] S160, perform frequency domain transformation on the second processed signal, and extract complex amplitude and phase information at non-power frequency excitation frequencies.

[0070] In one possible implementation, a frequency domain transformation is performed on the second processed signal, and complex amplitude and phase information at non-power frequency excitation frequencies are extracted. Specifically, this includes: inputting the second processed signal to a frequency domain transformation module, performing a fast Fourier transform operation to generate frequency domain complex spectrum data; the frequency domain transformation module calls a window function construction module to synchronously generate current window function parameters, ensuring the transformation interval is consistent with the window function structure and improving spectrum resolution accuracy; after acquiring the frequency domain complex spectrum data, the non-power frequency excitation frequency is input to a frequency positioning unit, which calculates the corresponding frequency index position based on the number of sampling points and the sampling frequency; the frequency index position is located in the frequency domain complex spectrum data, and the complex spectrum value at the corresponding position is extracted as the target complex amplitude and phase information; if the non-power frequency excitation frequency is not perfectly aligned with the frequency index, a frequency interpolation module is called to perform a spectral centroid method or a complex linear interpolation method to complete complex amplitude and phase interpolation between adjacent frequency points; the interpolated complex amplitude and phase information is output to a dielectric loss calculation module as input parameters for subsequent calculation of the dielectric loss factor using the complex current method.

[0071] Specifically, the second processed signal is input to the frequency domain transformation module, where a Fast Fourier Transform (FFT) operation is performed to generate complex spectrum data in the frequency domain. The second processed signal refers to the time-domain sampled data after weighting by a window function, possessing high spectral purity and sideband energy suppression capabilities. After inputting this signal into the frequency domain transformation module, a Fast Fourier Transform (FFT) operation is performed, which accelerates the discrete-time signal using DFT to extract the complex amplitude and phase information of each frequency component. The Fast Fourier Transform is an important algorithm for identifying frequency components in a time-domain signal. Its output is a complex spectrum sequence, representing the complex response at each frequency point, where the real part represents the cosine component and the imaginary part represents the sine component; together, they constitute the amplitude and phase characteristics at that frequency point.

[0072] The frequency domain transformation module calls the window function construction module to synchronously generate the current window function parameters, ensuring that the transformation interval is consistent with the window function structure and improving the accuracy of spectrum resolution. During frequency domain transformation, to avoid spectral leakage caused by window function truncation errors, it is necessary to ensure that the time series covered by the window function is completely consistent with the actual sampling interval participating in the Fourier transform. Therefore, the frequency domain transformation module needs to synchronously call the window function construction module to perform parameter matching and verification of the weighted window shape, sampling length, and position before the transformation, ensuring that the FFT input is the effective signal segment after time-domain windowing processing. This matching process is crucial for improving frequency domain resolution and avoiding boundary effects and phase drift, especially in scenarios where the target frequency and the main interference frequency are close together.

[0073] After acquiring the complex spectrum data in the frequency domain, the non-power frequency excitation frequency is input into the frequency positioning unit. The frequency positioning unit calculates the corresponding frequency index position based on the number of sampling points and the sampling frequency. The frequency index position refers to the array index in the complex spectrum array output by the FFT that is closest to the target non-power frequency excitation frequency. According to the definition of Fast Fourier Transform, the frequency index position... The corresponding actual frequency is ,in Sampling frequency, The number of FFT points. The frequency positioning unit is based on the target's non-power frequency excitation frequency. Inverse solution to find the index position Provided that the frequency deviation remains within a controllable range, this index will serve as a reference point for subsequent frequency domain interpolation or direct extraction.

[0074] In the frequency domain complex spectrum data, locate the frequency index position and extract the complex spectrum value at the corresponding position as the target complex amplitude and phase information. Extract the values ​​located at the corresponding positions from the FFT result array using indexing operations. complex values The real part is The imaginary part is The amplitude corresponding to that point can be calculated from this. , and phase This complex value represents the frequency response of the voltage or current signal at a non-power frequency excitation frequency, and is the core data for subsequent calculations of equivalent impedance and dielectric loss factor using the complex current method.

[0075] If the non-power frequency excitation frequency is not perfectly aligned with the frequency index, the frequency interpolation module is invoked to execute either the spectral centroid method or the complex linear interpolation method to perform complex amplitude and phase interpolation between adjacent frequency points. Since the non-power frequency excitation frequency is generally an adjustable floating-point value, and the FFT frequency resolution is limited, the target frequency often falls between the two indices. In this case, an interpolation algorithm is used to improve frequency domain accuracy. The spectral centroid method calculates the dominant frequency shift based on the power spectral density weighted by the three frequency points adjacent to the excitation frequency, and is suitable for high signal-to-noise ratio scenarios; the complex linear interpolation method... Linear interpolation of the complex spectral values ​​at both ends of the interval is performed using the following formula:

[0076] in and Frequency index and The complex spectral value at that location.

[0077] The interpolated complex amplitude and phase information are output to the dielectric loss calculation module as input parameters for subsequent calculation of the dielectric loss factor using the complex current method. After interpolation, the obtained complex amplitude and phase information represent the true frequency domain response at non-power frequency excitation frequencies, corresponding to the excitation voltage and response current, respectively. This pair of complex signals is organized into complex voltage and complex current forms and input to the dielectric loss calculation module, where it is used to construct a complex impedance. The dielectric loss factor is calculated by obtaining the ratio of the real to the imaginary parts of the complex impedance. Complete and high-precision complex amplitude and phase are prerequisites for ensuring the stability, sensitivity, and traceability of the dielectric loss factor. In this embodiment, when the sampling rate is 10kHz and the FFT points are 4096, the frequency resolution is approximately 2.44Hz, and the target frequency is 46.3Hz. A high-precision frequency domain response can be obtained by performing linear interpolation between adjacent indices 46Hz and 48.44Hz.

[0078] S170 calculates the dielectric loss factor of the current transformer based on the complex current method for the complex amplitude and phase information, and outputs the dielectric loss value of the current transformer.

[0079] In one possible implementation, the dielectric loss factor of the current transformer is calculated based on the complex current method for the complex amplitude and phase information, and the dielectric loss value of the current transformer is output. Specifically, this includes: determining the complex amplitude of the excitation voltage and the complex amplitude of the response current included in the complex amplitude, wherein the complex amplitude of the excitation voltage includes a first amplitude and a first phase, and the complex amplitude of the response current includes a second amplitude and a second phase; calculating the complex impedance value based on the complex amplitude of the excitation voltage and the complex amplitude of the response current, wherein the complex impedance value is obtained by dividing the complex value of the excitation voltage composed of the first amplitude and the first phase by the complex value of the response current composed of the second amplitude and the second phase; decomposing the complex impedance value into a real component and an imaginary component, wherein the real component corresponds to the resistive component and the imaginary component corresponds to the capacitive reactance component; using the ratio of the resistive component to the capacitive reactance component as the dielectric loss factor of the current transformer; and outputting the dielectric loss factor of the current transformer as the dielectric loss value of the current transformer.

[0080] Specifically, the complex amplitude includes the complex amplitude of the excitation voltage and the complex amplitude of the response current. The complex amplitude of the excitation voltage includes a first amplitude and a first phase, and the complex amplitude of the response current includes a second amplitude and a second phase. In this step, the complex amplitude of the excitation voltage refers to constructing the amplitude and phase of the excitation voltage extracted at the non-power frequency excitation frequency into a complex form, denoted as:

[0081] in For the first value, This is the first phase. The corresponding complex amplitude of the response current is denoted as... ,in For the second amplitude, This is the second phase. Both phases originate from the interpolation output of the aforementioned frequency domain extraction module, ensuring that the phase and amplitude information are aligned at the excitation frequency and that a unified amplitude-phase representation is maintained in the complex domain, providing a foundation for complex impedance calculation.

[0082] Based on the complex amplitude of the excitation voltage and the complex amplitude of the response current, the complex impedance value is calculated. The complex impedance value is obtained by dividing the complex value of the excitation voltage formed by the first amplitude and the first phase by the complex value of the response current formed by the second amplitude and the second phase. Specifically, the complex excitation voltage... With complex response current Substituting into the complex form of Ohm's law, calculate the complex impedance:

[0083] This complex impedance value contains both magnitude and phase difference information, and can fully reflect the equivalent impedance behavior of the current transformer at the excitation frequency, where the phase difference... It describes the amount by which voltage leads or lags current.

[0084] The complex impedance is decomposed into real and imaginary components. The real components correspond to the resistive component, and the imaginary components correspond to the capacitive reactance component. The complex impedance is expressed in complex form in a rectangular coordinate system. ,in The real part represents the resistance component corresponding to the active power. The imaginary part represents the capacitive reactance component corresponding to reactive power. Through complex factorization, the energy consumption (represented by resistance) and energy storage (represented by capacitive reactance) behavior of the current transformer insulation system can be accurately quantified, thereby distinguishing between normal insulation and degradation trends.

[0085] The ratio of the resistive component to the capacitive reactance component is used as the dielectric loss factor of the current transformer. The dielectric loss factor of a current transformer is defined as the ratio of the resistive component to the capacitive reactance component, denoted as . ,in The dielectric loss factor is the ratio of the dielectric loss angle to the reactive power lost per unit electric field. This ratio quantifies the dissipation capacity of the insulating dielectric and is equivalent to the ratio of active power to reactive power lost by the dielectric under a unit electric field. It is a key parameter characterizing the degree of insulation performance degradation of a current transformer. Under normal operating conditions, the dielectric loss factor should be stable within the millisecond range. If it exceeds this range, it may indicate a significant abnormality in energy dissipation during dielectric polarization.

[0086] The dielectric loss factor of the current transformer is output as the dielectric loss value of the current transformer. The final dielectric loss value of the current transformer is the calculated dielectric loss factor, which serves as an important input parameter for equipment condition assessment, insulation degradation trend analysis, and online early warning systems. In engineering applications, when the complex impedance of a certain type of current transformer at an excitation frequency of 47Hz is... Ohm, then the dielectric loss factor is By comparing historical data, it can be determined that the insulation is in the early stage of aging and should be included in the key monitoring targets for operation and maintenance. This output supports the measurement system in forming a time series database, providing a quantitative basis for long-term trend prediction and insulation life estimation.

[0087] This embodiment also discloses a current transformer dielectric loss anti-interference measurement device based on a frequency converter power supply, referring to... Figure 2 The device includes an acquisition module 201, a processing module 202, and an output module 203. It is used to execute any of the above-described methods for measuring the dielectric loss and interference immunity of a current transformer based on a frequency converter power supply, wherein: The acquisition module 201 is used to acquire the excitation path between the primary winding of the current transformer and ground, and dynamically identify the main interference frequency in the measurement field based on the frequency tracking algorithm to construct the frequency offset parameters.

[0088] The processing module 202 is used to dynamically generate a pair of non-power frequency excitation frequencies based on the frequency offset parameters, and control the output of the variable frequency power supply as a sine wave excitation signal based on the non-power frequency excitation frequencies.

[0089] The processing module 202 is used to input the sinusoidal excitation signal into the current transformer measurement circuit, and then simultaneously sample the excitation voltage signal and the response current signal to obtain the sampled signal.

[0090] The processing module 202 is used to adjust the notch frequency of the sampled signal based on the frequency offset parameter to filter out the main interference component and obtain the first processed signal.

[0091] The processing module 202 is used to dynamically construct a window function based on the non-power frequency excitation frequency and frequency band characteristics, and to weight and process the first processed signal to obtain the second processed signal.

[0092] The processing module 202 is used to perform frequency domain transformation on the second processed signal and extract complex amplitude and phase information at non-power frequency excitation frequencies.

[0093] The output module 203 is used to calculate the dielectric loss factor of the current transformer based on the complex current method, and output the dielectric loss value of the current transformer, based on the complex amplitude and phase information.

[0094] In one possible implementation, the acquisition module 201 is used to apply an excitation signal between the primary winding of the current transformer and ground to establish an excitation channel.

[0095] The acquisition module 201 is used to acquire the background response signal in the excitation channel. The background response signal includes the original sampled data of the voltage signal and the current signal.

[0096] The processing module 202 is used to perform spectrum extraction on the background response signal, identify the frequency component with the largest amplitude in the background response signal, and determine the main interference frequency.

[0097] The processing module 202 is used to perform a difference calculation between the main interference frequency and the preset ideal power frequency to generate frequency offset parameters.

[0098] In one possible implementation, the processing module 202 is used to select a pair of target non-power frequency excitation frequencies with the smallest overlap with the main interference frequency and harmonic frequency band from a set of multiple non-power frequency excitation frequencies in a preset frequency candidate set according to the frequency offset parameter, wherein the frequency difference between the multiple non-power frequency excitation frequencies satisfies the frequency domain decoupling condition and the system frequency resolution constraint.

[0099] The processing module 202 is used to input the target non-power frequency excitation frequency into the waveform modulation module and control the frequency converter to call the waveform parameters corresponding to the target non-power frequency excitation frequency.

[0100] The output module 203 is used to generate a pair of sinusoidal excitation signals with stable frequency, consistent amplitude and synchronized phase by the waveform modulation module according to the waveform parameters.

[0101] In one possible implementation, the processing module 202 is used to input the sinusoidal excitation signal through the isolation drive unit to the excitation channel formed between the primary winding of the current transformer and ground.

[0102] The processing module 202 is used to obtain the response current signal through the current detection device based on the fact that the secondary winding terminals of the current transformer are short-circuited and grounded to form a complete closed measurement circuit.

[0103] The processing module 202 is used to acquire the excitation voltage signal by means of a voltage sensor, based on the voltage sampling channels arranged at both ends of the excitation channel.

[0104] The processing module 202 is used to perform parallel analog-to-digital conversion on the excitation voltage signal and the response current signal respectively, and construct the excitation voltage signal and the response current signal after analog-to-digital conversion into a data structure aligned with the time sequence to form a sampling signal.

[0105] In one possible implementation, the processing module 202 is used to construct a multidimensional cost function based on a family of cosine lifting window functions for the first processed signal. The multidimensional cost function aims to maximize the spectral purity of the excitation frequency and minimize the energy leakage of adjacent frequency bands.

[0106] Processing module 202 is used to output frequency band characteristic parameters based on the non-power frequency excitation frequency. The frequency band characteristic parameters include the frequency density and the distribution of the main interference harmonics on both sides of the non-power frequency excitation frequency.

[0107] The processing module 202 is used to execute a multi-objective parameter optimization algorithm to generate an optimal window function sample sequence based on the non-power frequency excitation frequency and frequency band characteristic parameters.

[0108] The processing module 202 is used to input the window function sample sequence into the multidimensional cost function, perform a time-domain point-to-point weighting operation on the first processed signal, and output the second processed signal.

[0109] In one possible implementation, the processing module 202 is used to perform point-by-point multiplication on the weight coefficients at corresponding positions in the window function sample sequence for each sampling point in the first processed signal, and to construct a weighted result sequence.

[0110] The processing module 202 is used to call the frequency density gradient mapping function to perform dynamic edge correction on the weighted result sequence during the point-by-point multiplication process. The frequency density gradient mapping function is generated based on the energy distribution trend of the excitation frequency and the frequency adjacent to the frequency in the frequency band characteristic parameters.

[0111] The processing module 202 is used to take the weighted result sequence after dynamic edge correction as the second processing signal.

[0112] In one possible implementation, the processing module 202 is configured to determine the complex amplitude of the excitation voltage and the complex amplitude of the response current included in the complex amplitude, wherein the complex amplitude of the excitation voltage includes a first amplitude and a first phase, and the complex amplitude of the response current includes a second amplitude and a second phase.

[0113] The processing module 202 is used to calculate the complex impedance value based on the complex amplitude of the excitation voltage and the complex amplitude of the response current. The complex impedance value is obtained by dividing the complex value of the excitation voltage formed by the first amplitude and the first phase by the complex value of the response current formed by the second amplitude and the second phase.

[0114] The processing module 202 is used to decompose the complex impedance value into real components and imaginary components, with the real components corresponding to the resistive components and the imaginary components corresponding to the capacitive reactance components.

[0115] Processing module 202 is used to use the ratio of resistive component to capacitive reactance component as the dielectric loss factor of current transformer.

[0116] The processing module 202 is used to output the dielectric loss factor of the current transformer as the dielectric loss value of the current transformer.

[0117] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0118] This embodiment also discloses an electronic device, as shown in the reference. Figure 3 The electronic device may include: at least one processor 301, at least one communication bus 302, user interface 303, network interface 304, and at least one memory 305.

[0119] The communication bus 302 is used to enable communication between these components.

[0120] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0121] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0122] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 305, and by calling data stored in memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications. The GPU is responsible for rendering and drawing the content required for display. The modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0123] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory may include a non-transitory computer-readable storage medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), and instructions for implementing the various method embodiments described above. The data storage area may store data involved in the various method embodiments described above. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. As a computer storage medium, the memory 305 may include an operating system, a network communication module, a user interface 303 module, and an application program for a current transformer dielectric loss anti-interference measurement method based on a frequency converter power supply.

[0124] exist Figure 3 In the illustrated electronic device, the user interface 303 is primarily used to provide an input interface for the user and to acquire user-input data. The processor 301 can be used to call an application program stored in the memory 305, which represents a method for measuring the dielectric loss and interference immunity of a current transformer based on a frequency converter. When executed by one or more processors 301, the electronic device performs one or more methods as described in the above embodiments.

[0125] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0126] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0128] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0129] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0130] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory 305 and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory 305 includes various media capable of storing program code, such as a USB flash drive, external hard drive, magnetic disk, or optical disk.

[0131] This application also discloses a computer-readable storage medium storing instructions. When executed by one or more processors 301, these instructions cause an electronic device to perform one or more methods as described in the above embodiments.

[0132] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of other embodiments of this disclosure upon considering the specification and the disclosure of practical truths. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for measuring the dielectric loss and interference immunity of a current transformer based on a variable frequency power supply, characterized in that, The method includes: Obtain the excitation path between the primary winding of the current transformer and ground, and dynamically identify the main interference frequency in the measurement field based on the frequency tracking algorithm to construct the frequency offset parameter; A pair of non-power frequency excitation frequencies are dynamically generated based on the frequency offset parameters, and the frequency converter is controlled to output a sinusoidal excitation signal based on the non-power frequency excitation frequencies. After the sinusoidal excitation signal is input into the current transformer measurement circuit, the excitation voltage signal and the response current signal are sampled simultaneously to obtain the sampled signal. The sampling signal is adjusted based on the frequency offset parameter to remove the main interference component, thereby obtaining the first processed signal; Based on the non-power frequency excitation frequency and bandwidth characteristics, a window function is dynamically constructed and the first processed signal is weighted and processed to obtain the second processed signal; The second processed signal is subjected to frequency domain transformation, and the complex amplitude and phase information at the non-power frequency excitation frequency is extracted; Based on the complex amplitude and phase information, the dielectric loss factor of the current transformer is calculated using the complex current method, and the dielectric loss value of the current transformer is output.

2. The method for measuring the dielectric loss and anti-interference of a current transformer based on a frequency converter power supply according to claim 1, characterized in that, The process of acquiring the excitation path between the primary winding of the current transformer and ground, and dynamically identifying the main interference frequency in the measurement field based on a frequency tracking algorithm to construct frequency offset parameters, specifically includes: An excitation signal is applied between the primary winding of the current transformer and ground to establish the excitation channel; The background response signal in the excitation channel is acquired, and the background response signal includes the original sampled data of voltage signal and current signal; Spectrum extraction is performed on the background response signal to identify the frequency component with the largest amplitude in the background response signal and determine it as the main interference frequency; The frequency offset parameter is generated by performing a difference calculation between the main interference frequency and the preset ideal power frequency.

3. The method for measuring the dielectric loss and anti-interference of a current transformer based on a frequency converter power supply according to claim 1, characterized in that, The step of dynamically generating a pair of non-power frequency excitation frequencies based on the frequency offset parameters, and controlling the output of a sinusoidal excitation signal from the frequency converter based on the non-power frequency excitation frequencies, specifically includes: Based on the frequency offset parameter, among multiple non-power frequency excitation frequencies in the preset frequency candidate set, a pair of target non-power frequency excitation frequencies with the smallest overlap with the main interference frequency and harmonic frequency band are selected, and the frequency difference between the multiple non-power frequency excitation frequencies satisfies the frequency domain decoupling condition and the system frequency resolution constraint. The target non-power frequency excitation frequency is input into the waveform modulation module to control the frequency converter to call the waveform parameters corresponding to the target non-power frequency excitation frequency; Based on the waveform parameters, the waveform modulation module generates a pair of sinusoidal excitation signals with stable frequency, consistent amplitude, and synchronized phase.

4. The method for measuring the dielectric loss and anti-interference of a current transformer based on a frequency converter power supply according to claim 1, characterized in that, After inputting the sinusoidal excitation signal into the current transformer measurement circuit, the excitation voltage signal and the response current signal are simultaneously sampled to obtain the sampled signal, specifically including: The sinusoidal excitation signal is input to the excitation channel formed between the primary winding of the current transformer and ground through the isolation drive unit; The response current signal is obtained by shorting and grounding the secondary winding terminals of the current transformer to form a complete closed measurement circuit, and then using a current detection device. By arranging voltage sampling channels at both ends of the excitation channel, the excitation voltage signal is acquired through a voltage sensor. The excitation voltage signal and the response current signal are respectively subjected to parallel analog-to-digital conversion. The excitation voltage signal and the response current signal after analog-to-digital conversion are constructed into a data structure aligned with the time sequence to form the sampling signal.

5. The method for measuring the dielectric loss and anti-interference of a current transformer based on a frequency converter power supply according to claim 1, characterized in that, The step of dynamically constructing a window function based on the non-power frequency excitation frequency and bandwidth characteristics, and then weighting and processing the first processed signal to obtain the second processed signal, specifically includes: The first processed signal is used to construct a multidimensional cost function based on a family of cosine lifting window functions. The multidimensional cost function aims to maximize the spectral purity of the excitation frequency and minimize the energy leakage of adjacent frequency bands. According to the non-power frequency excitation frequency output frequency band characteristic parameters, the frequency band characteristic parameters include the frequency density and main interference harmonic distribution on both sides of the non-power frequency excitation frequency; A multi-objective parameter optimization algorithm is executed to generate an optimal window function sample sequence based on the non-power frequency excitation frequency and the frequency band characteristic parameters; The window function sample sequence is input into the multidimensional cost function, and a time-domain point-to-point weighting operation is performed on the first processed signal to output the second processed signal.

6. The method for measuring the dielectric loss and anti-interference of a current transformer based on a frequency converter power supply according to claim 5, characterized in that, The step of inputting the window function sample sequence into the multidimensional cost function, performing a time-domain point-to-point weighting operation on the first processed signal, and outputting the second processed signal specifically includes: Perform point-by-point multiplication on the weight coefficients at corresponding positions in the window function sample sequence for each sampling point in the first processed signal to construct a weighted result sequence; During the point-by-point product process, the frequency density gradient mapping function is called to perform dynamic edge correction on the weighted result sequence. The frequency density gradient mapping function is generated based on the energy distribution trend of the excitation frequency in the frequency band characteristic parameters. The weighted result sequence after dynamic edge correction is used as the second processing signal.

7. The method for measuring the dielectric loss and anti-interference of a current transformer based on a frequency converter power supply according to claim 1, characterized in that, The calculation of the dielectric loss factor of the current transformer based on the complex current method, using the complex amplitude and phase information, and the output of the dielectric loss value of the current transformer, specifically includes: The complex amplitude is determined to include the complex amplitude of the excitation voltage and the complex amplitude of the response current, wherein the complex amplitude of the excitation voltage includes a first amplitude and a first phase, and the complex amplitude of the response current includes a second amplitude and a second phase; The complex impedance value is calculated based on the complex amplitude of the excitation voltage and the complex amplitude of the response current. The complex impedance value is obtained by dividing the complex value of the excitation voltage formed by the first amplitude and the first phase by the complex value of the response current formed by the second amplitude and the second phase. The complex impedance value is decomposed into a real component and an imaginary component, where the real component corresponds to the resistive component and the imaginary component corresponds to the capacitive reactance component. The ratio of the resistive component to the capacitive reactance component is used as the dielectric loss factor of the current transformer. The dielectric loss factor of the current transformer is output as the dielectric loss value of the current transformer.

8. A current transformer dielectric loss anti-interference measurement device based on a variable frequency power supply, characterized in that, The device is used to perform a current transformer dielectric loss anti-interference measurement method based on a frequency converter power supply as described in any one of claims 1-7. The device includes an acquisition module (201), a processing module (202), and an output module (203), wherein: The acquisition module (201) is used to acquire the excitation path between the primary winding of the current transformer and the ground, and to dynamically identify the main interference frequency in the measurement site based on the frequency tracking algorithm, and to construct the frequency offset parameter. The processing module (202) is used to dynamically generate a pair of non-power frequency excitation frequencies according to the frequency offset parameters, and control the frequency converter to output a sine wave excitation signal according to the non-power frequency excitation frequencies. The processing module (202) is used to input the sinusoidal excitation signal into the current transformer measurement circuit, and then simultaneously sample the excitation voltage signal and the response current signal to obtain the sampled signal. The processing module (202) is used to adjust the notch frequency of the sampled signal based on the frequency offset parameter to filter out the main interference component, thereby obtaining a first processed signal; The processing module (202) is used to dynamically construct a window function and weight the first processed signal according to the non-power frequency excitation frequency and frequency band characteristics to obtain the second processed signal; The processing module (202) is used to perform frequency domain transformation on the second processed signal and extract the complex amplitude and phase information at the non-power frequency excitation frequency; The output module (203) is used to calculate the dielectric loss factor of the current transformer based on the complex amplitude and the phase information, and output the dielectric loss value of the current transformer.

9. An electronic device, characterized in that, The device includes a processor (301), a communication bus (302), a user interface (303), a network interface (304), and a memory (305). The memory (305) is used to store instructions. The user interface (303) and the network interface (304) are both used to communicate with other devices. The communication bus (302) is used to realize the connection and communication between the components within the electronic device. The processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device performs the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-7.

Images