CN-122017501-A - Insulation detection method and system for power equipment

CN122017501ACN 122017501 ACN122017501 ACN 122017501ACN-122017501-A

Abstract

The invention provides an insulation detection method and system of power equipment, wherein the method comprises the steps of injecting detection pulse signals into a power supply loop and collecting mixed response signals; the method comprises the steps of carrying out time-frequency domain joint analysis on a mixed response signal, separating and extracting frequency band characteristic parameters of a dynamic noise signal in a current working period, including noise center frequency drift amount, harmonic energy peak value and frequency spectrum coverage bandwidth, obtaining a filtering algorithm corresponding to a current frequency conversion working condition through similarity matching, generating real-time dynamic filtering parameters, enabling the center of a stop band and the width of the stop band to be adaptively tracked and updated along with the noise center frequency drift amount and the frequency spectrum coverage bandwidth, filtering the mixed response signal based on the dynamic filtering parameters and the filtering algorithm to obtain a reconstruction signal aiming at an insulation leakage current signal, and calculating a real-time insulation resistance value according to the amplitude and the phase of the reconstruction signal. The method and the device realize self-adaptive tracking and filtering of dynamic noise, and remarkably improve the detection precision of the insulation resistance under the complex frequency conversion working condition.

Inventors

ZHANG CHUNMAO
ZHONG JIN
DENG HUALIANG
HUANG JIANJIANG
Yuan Gongxian
ZHOU BO
LIU DINGPING
LIU XIE
ZHOU JUN
XU MIN
ZHENG GUO
LI YUXUAN

Assignees

湛江电力有限公司
广东埃森环保科技有限公司
西安凯锐电气技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260407

Claims (10)

1. A method of detecting insulation of an electrical device, comprising: Injecting a detection pulse signal into a power supply loop of target power equipment, and collecting a mixed response signal of the power supply loop, wherein the mixed response signal comprises an insulation leakage current signal excited by the detection pulse signal and a dynamic noise signal generated by the operation of frequency conversion equipment; Performing joint analysis of a time domain and a frequency domain on the mixed response signal, and separating and extracting frequency band characteristic parameters of the dynamic noise signal in the current working period according to a joint analysis result, wherein the frequency band characteristic parameters comprise noise center frequency drift amount, harmonic energy peak value and frequency spectrum coverage bandwidth; The frequency band characteristic parameters and a pre-constructed variable frequency interference characteristic library are subjected to similarity matching to obtain a filtering algorithm corresponding to the current variable frequency working condition, and real-time dynamic filtering parameters are generated based on the filtering algorithm and the noise center frequency drift amount, wherein the stop band center and the stop band width of the dynamic filtering parameters are subjected to self-adaptive tracking updating along with the noise center frequency drift amount and the frequency spectrum coverage bandwidth; filtering the mixed response signal based on the dynamic filtering parameter and the filtering algorithm to filter the dynamic noise signal overlapped with the frequency band of the detection pulse signal, so as to obtain a reconstruction signal for an insulation leakage current signal; And according to the amplitude and the phase of the reconstruction signal, the real-time insulation resistance value of the power supply loop of the target power equipment in the running state is calculated by combining the voltage parameter of the detection pulse signal, so as to realize insulation detection.
2. The insulation detection method of an electrical device according to claim 1, wherein the performing a joint analysis of a time domain and a frequency domain on the mixed response signal, and separating and extracting a frequency band characteristic parameter of the dynamic noise signal in a current working period according to a joint analysis result, includes: Adopting synchronous compression wavelet transformation to the mixed response signals, and carrying out compression rearrangement on signal energy to instantaneous frequency points on a frequency domain to obtain a time-frequency energy distribution matrix; based on the time-frequency energy distribution matrix, extracting an energy ridge line track of the dynamic noise signal on a time-frequency plane by using an energy gradient-based ridge line extraction algorithm; calculating the difference between the maximum value and the minimum value of the instantaneous frequency of the energy ridge line track in the current working period as the noise center frequency drift amount of the dynamic noise signal; Taking the energy ridge line track as a central axis, and performing layer-by-layer energy integration on two sides of the time-frequency energy distribution matrix along the frequency axis to obtain accumulated integrated energy; And when the accumulated integral energy reaches a preset proportional threshold of total noise energy at the current moment, acquiring an upper frequency boundary and a lower frequency boundary corresponding to the current moment, and taking a difference value between the upper frequency boundary and the lower frequency boundary as the frequency spectrum coverage bandwidth.
3. The insulation detection method of power equipment according to claim 1, wherein the variable frequency interference feature library stores a plurality of standard working condition modes and an adaptive filtering function template stored in association with each standard working condition mode; and matching the frequency band characteristic parameters with a pre-constructed variable frequency interference characteristic library in a similarity mode to obtain a filtering algorithm corresponding to the current variable frequency working condition, wherein the filtering algorithm comprises the following steps: characterizing the frequency band characteristic parameters as multidimensional working condition characteristic vectors of the current noise distribution characteristics; Projecting the multidimensional working condition feature vector to a preset high-dimensional feature space to obtain a projection result, and respectively calculating weighted Markov distances between the projection result and the centers of all standard working condition modes; Selecting a standard working condition mode with the minimum weighted mahalanobis distance as a target matching mode, and calling a self-adaptive filtering function template corresponding to the target matching mode; And determining the order and the topological structure of a filter in the self-adaptive filtering function template according to the real-time change rate of the noise center frequency drift amount, so as to generate a filtering algorithm aiming at the current frequency conversion working condition based on the order and the topological structure.
4. The insulation detection method of a power device according to claim 1, wherein the filtering the mixed response signal based on the dynamic filtering parameter and the filtering algorithm to filter the dynamic noise signal overlapping the frequency band of the detection pulse signal, to obtain a reconstructed signal for an insulation leakage current signal, includes: Constructing a time-varying notch filter according to the dynamic filtering parameters and the filtering algorithm; inputting the mixed response signal into the time-varying notch filter to obtain a preliminary filtering signal; Rearranging the preliminary filtering signals according to the time reverse order to obtain a rearranging result; Inputting the rearrangement result into the time-varying notch filter for secondary filtering treatment to obtain a zero phase shift denoising signal; the detection pulse signal is taken as a reference signal to construct a dynamic orthogonal reference coordinate system consisting of an inphase axis and an orthogonal axis, and the zero-phase removal noise signal is projected into the dynamic orthogonal reference coordinate system to obtain an inphase component and a quadrature component which have the same frequency as the detection pulse signal; and carrying out vector synthesis and amplitude gain compensation calculation by utilizing the in-phase component and the quadrature component aiming at the overlapping area of the frequency bands to obtain the reconstruction signal.
5. The insulation detection method of an electric power device according to claim 1, wherein the injecting the detection pulse signal into the power supply loop of the target electric power device and collecting the mixed response signal of the power supply loop further comprises: The frequency and amplitude of the detection pulse signal are adjusted according to the system-to-ground distributed capacitance of the power supply loop and the insulation resistance value obtained in the last detection period; And monitoring the system distributed capacitance to the ground, and when the system distributed capacitance to the ground is increased, reducing the frequency of the detection pulse signal and correspondingly increasing the output amplitude of the detection pulse signal so as to ensure that the signal-to-noise ratio of the mixed response signal is within a preset threshold range.
6. The insulation detection method of an electric power apparatus according to claim 1, further comprising, before the injection of the detection pulse signal to the power supply circuit of the target electric power apparatus: determining a grounding system type of the target power equipment; If the grounding system type is a TN system, monitoring the running state of the target power equipment; When the target power equipment is in an operation state, sending a standby enabling command to an insulation monitoring terminal through a MODBUS communication protocol to enable the insulation monitoring terminal to enter a standby state; when the target power equipment is monitored to exit the running state, controlling the insulation monitoring terminal to be put into operation so as to implement offline online monitoring; and if the grounding system type is an IT system, controlling the insulation monitoring terminal to directly enter a real-time online monitoring mode.
7. The insulation detection method of an electrical device according to claim 1, further comprising, after the calculating the real-time insulation resistance value of the power supply loop of the target electrical device in the operating state: Comparing the real-time insulation resistance with a preset first early warning threshold value and a preset second alarm threshold value respectively; When the real-time insulation resistance value is lower than a first early warning threshold value or a second early warning threshold value, fault positioning is carried out according to the polarity characteristic or the phase characteristic of the reconstruction signal; If the power supply loop is a direct current system, judging and displaying that an insulation fault occurs at the positive electrode or the negative electrode according to the current flow direction of the reconstruction signal; If the power supply loop is a three-phase alternating current system, the phase correlation between the reconstruction signal and each phase voltage is analyzed, and the phase line with the insulation fault is judged and displayed.
8. The insulation detection system of the power equipment is characterized by comprising an acquisition module, an extraction module, a filtering parameter generation module, a reconstruction module and an insulation detection module, wherein, The acquisition module is used for injecting a detection pulse signal into a power supply loop of target power equipment and acquiring a mixed response signal of the power supply loop, wherein the mixed response signal comprises an insulation leakage current signal excited by the detection pulse signal and a dynamic noise signal generated by the operation of frequency conversion equipment; the extraction module is used for carrying out joint analysis on the time domain and the frequency domain of the mixed response signal, and separating and extracting frequency band characteristic parameters of the dynamic noise signal in the current working period according to the joint analysis result, wherein the frequency band characteristic parameters comprise noise center frequency drift amount, harmonic energy peak value and frequency spectrum coverage bandwidth; The filtering parameter generation module is used for carrying out similarity matching on the frequency band characteristic parameters and a pre-constructed variable frequency interference characteristic library to obtain a filtering algorithm corresponding to the current variable frequency working condition, and generating real-time dynamic filtering parameters based on the filtering algorithm and the noise center frequency drift amount, wherein the stop band center and the stop band width of the dynamic filtering parameters carry out self-adaptive tracking updating along with the noise center frequency drift amount and the frequency spectrum coverage bandwidth; the reconstruction module is used for carrying out filtering processing on the mixed response signal based on the dynamic filtering parameter and the filtering algorithm so as to filter the dynamic noise signal overlapped with the frequency band of the detection pulse signal and obtain a reconstruction signal for an insulation leakage current signal; The insulation detection module is used for calculating the real-time insulation resistance value of the power supply loop of the target power equipment in the running state according to the amplitude and the phase of the reconstruction signal and the voltage parameter of the detection pulse signal so as to realize insulation detection.
9. The insulation detection system of an electrical device according to claim 8, wherein the extracting module performs a joint analysis of a time domain and a frequency domain on the mixed response signal, and separates and extracts the frequency band characteristic parameter of the dynamic noise signal in the current working period according to a joint analysis result, and the method comprises: the extraction module adopts synchronous compression wavelet transformation to the mixed response signals, and compresses and rearranges signal energy to instantaneous frequency points on a frequency domain to obtain a time-frequency energy distribution matrix; based on the time-frequency energy distribution matrix, extracting an energy ridge line track of the dynamic noise signal on a time-frequency plane by using an energy gradient-based ridge line extraction algorithm; calculating the difference between the maximum value and the minimum value of the instantaneous frequency of the energy ridge line track in the current working period as the noise center frequency drift amount of the dynamic noise signal; Taking the energy ridge line track as a central axis, and performing layer-by-layer energy integration on two sides of the time-frequency energy distribution matrix along the frequency axis to obtain accumulated integrated energy; And when the accumulated integral energy reaches a preset proportional threshold of total noise energy at the current moment, acquiring an upper frequency boundary and a lower frequency boundary corresponding to the current moment, and taking a difference value between the upper frequency boundary and the lower frequency boundary as the frequency spectrum coverage bandwidth.
10. The insulation detection system of power equipment according to claim 8, wherein the variable frequency interference feature library stores a plurality of standard working condition modes, and an adaptive filter function template stored in association with each standard working condition mode; the filtering parameter generating module performs similarity matching on the frequency band characteristic parameter and a pre-constructed variable frequency interference characteristic library to obtain a filtering algorithm corresponding to the current variable frequency working condition, and the filtering algorithm comprises the following steps: The filtering parameter generation module characterizes the frequency band characteristic parameter as a multidimensional working condition characteristic vector of the current noise distribution characteristic; Projecting the multidimensional working condition feature vector to a preset high-dimensional feature space to obtain a projection result, and respectively calculating weighted Markov distances between the projection result and the centers of all standard working condition modes; Selecting a standard working condition mode with the minimum weighted mahalanobis distance as a target matching mode, and calling a self-adaptive filtering function template corresponding to the target matching mode; And determining the order and the topological structure of a filter in the self-adaptive filtering function template according to the real-time change rate of the noise center frequency drift amount, so as to generate a filtering algorithm aiming at the current frequency conversion working condition based on the order and the topological structure.

Description

Insulation detection method and system for power equipment Technical Field The present invention relates to the field of power equipment detection, and in particular, to a method and a system for detecting insulation of power equipment. Background Along with the rapid development of a power system and the improvement of the automation level of industrial production, the safe and stable operation of the power supply loop equipment for the factory is important for continuous production. Insulation problems of electrical lines and equipment are significant safety hazards, and once the insulation performance is degraded, serious threats are caused to the stability and continuity of power supply. At present, the maintenance of power supply loop equipment in the power industry generally adopts offline manual inspection and periodic preventive tests. However, the conventional manual inspection method has the problems of long detection period, large workload and easy occurrence of missed detection and false detection, and cannot continuously and timely reflect the dynamic insulation condition of the equipment in the running state. In order to overcome the limitation of off-line detection, an on-line insulation monitoring technology is gradually applied, and by actively injecting alternating current or low-frequency pulse signals into a detected system and measuring response current, the on-line monitoring of insulation resistance is realized, and the insulation maintenance of equipment is pushed to develop from a preventive test to a predictive maintenance direction. The existing online insulation detection technology generally adopts a fixed frequency signal injection method, and is matched with a traditional hardware low-pass/band-pass filter or a fixed parameter software filtering algorithm to extract leakage current signals of insulation response. However, in modern industrial production sites, a large number of variable frequency devices inject broadband and dynamically varying harmonic noise into the grid during operation. The noise spectrum generated by the frequency converter can be subjected to irregular dynamic drift due to the real-time change of the working frequency and the load of the frequency converter. When the prior art is applied to such working conditions, the mismatch between the fixed and static filtering mechanisms and the dynamic drift noise can cause serious distortion of the finally extracted response signals, so that a larger deviation exists between the detection result of the insulation resistance and the actual value, frequent false alarms of a monitoring system are extremely easy to be caused in the actual application, and the safety and reliability of the production operation of a power plant are seriously influenced. Disclosure of Invention The invention provides an insulation detection method and an insulation detection system for power equipment, which aim to solve the technical problem of how to reduce the deviation between an insulation resistance detection result and an actual value. In order to solve the above technical problems, the present application provides an insulation detection method for an electrical device, including: Injecting a detection pulse signal into a power supply loop of target power equipment, and collecting a mixed response signal of the power supply loop, wherein the mixed response signal comprises an insulation leakage current signal excited by the detection pulse signal and a dynamic noise signal generated by the operation of frequency conversion equipment; Performing joint analysis of a time domain and a frequency domain on the mixed response signal, and separating and extracting frequency band characteristic parameters of the dynamic noise signal in the current working period according to a joint analysis result, wherein the frequency band characteristic parameters comprise noise center frequency drift amount, harmonic energy peak value and frequency spectrum coverage bandwidth; The frequency band characteristic parameters and a pre-constructed variable frequency interference characteristic library are subjected to similarity matching to obtain a filtering algorithm corresponding to the current variable frequency working condition, and real-time dynamic filtering parameters are generated based on the filtering algorithm and the noise center frequency drift amount, wherein the stop band center and the stop band width of the dynamic filtering parameters are subjected to self-adaptive tracking updating along with the noise center frequency drift amount and the frequency spectrum coverage bandwidth; filtering the mixed response signal based on the dynamic filtering parameter and the filtering algorithm to filter the dynamic noise signal overlapped with the frequency band of the detection pulse signal, so as to obtain a reconstruction signal for an insulation leakage current signal; And according to the amplitude and the phase of the reconstruction signal