Partial discharge on-line early warning device and method for 35kV high-voltage switch cabinet

Abstract

The invention relates to a partial discharge on-line early warning device and method for a 35kV high-voltage switch cabinet, and belongs to the technical field of early warning devices.The partial discharge on-line early warning device comprises an acquisition module, a receiving module and a master station module, and data acquired by the acquisition module is wirelessly transmitted to the receiving module; the receiving module transmits data to the master station module through the RS485 module, the acquisition module comprises a plurality of wireless temperature sensors, a wireless IMD sensor unit and a high-frequency current sensor, and the high-frequency current sensor is electrically connected with the wireless IMD sensor unit. The wireless IMD sensor unit and the high-frequency current sensor are detachably installed in a high-voltage cable bin of the high-voltage switch cabinet. According to the partial discharge on-line early warning device for the 35kV high-voltage switch cabinet, the temperature sensor is arranged to monitor the temperature of the switch cabinet, find out thermal fault hidden dangers in advance, give an alarm and monitor discharge signals of the switch cabinet, so that the reliability of safe operation of the high-voltage switch cabinet is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of early warning devices, in particular to a partial discharge online early warning device and method for a 35kV high-voltage switch cabinet. BACKGROUND

[0002] The high-voltage switch cabinet is an important link for ensuring the operation of the power grid. Due to the small internal space, numerous parts, complex structure and small insulation distance of the switch cabinet, it is more prone to insulation defects than other power equipment, thereby bringing great hidden dangers to the safe operation of the equipment. Defects such as air gap, impurities and spikes in the insulation of electrical equipment make the electric field distribution in the insulation of the switch cabinet uneven under the action of a strong electric field, and the electric field strength at the defect site will increase, thereby easily causing a discharge that does not penetrate the entire insulation, i.e. partial discharge. Therefore, monitoring and testing the partial discharge of electrical equipment is an important means for evaluating the insulation condition of the equipment, and is one of the effective measures for discovering latent faults of the equipment, realizing fault early warning and avoiding faults.

[0003] During the long-term operation of the switch cabinet, due to ground settlement and vibration, impact vibration during operation, the influence of electric power, oxidation of the surrounding air, etc., poor contact may occur between the high-voltage conductors of the switch cabinet, resulting in an increase in resistance, which will cause overheating when the load current passes through. Overheating of the live part will cause insulation aging or even breakdown, thereby causing a short circuit and forming a major accident, causing huge economic losses and threatening personal safety. Therefore, a partial discharge online early warning device for a high-voltage switch cabinet is needed. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides a partial discharge online early warning device for a 35kV high-voltage switch cabinet to monitor the temperature of the switch cabinet, discover thermal fault hidden dangers in advance and make an alarm, monitor the discharge signal of the switch cabinet and improve the reliability of the safe operation of the high-voltage switch cabinet.

[0005] In order to achieve the above-mentioned application purposes, the present application provides, in one aspect, a partial discharge online early warning device for a 35kV high-voltage switch cabinet, comprising a collection module, a receiving module and a master station module. The data collected by the collection module is wirelessly transmitted to the receiving module. The receiving module transmits the data to the master station module through an RS485 module. The collection module comprises a plurality of wireless temperature sensors, a wireless IMD sensor unit and a high-frequency current sensor. The high-frequency current sensor is electrically connected to the wireless IMD sensor unit. The wireless IMD sensor unit and the high-frequency current sensor are detachably installed in the high-voltage cable compartment of the high-voltage switch cabinet.

[0006] Further, the measurement range of the USS measurement of the wireless IMD sensor unit is 0-60dB, the resolution is 1dB, the accuracy is ±1dB, and the center frequency is 40KHz. The linearity error of the TEV measurement of the wireless IMD sensor unit is within ±10%, the working frequency band is 3-100MHz, and the humidity detection range of the humidity measurement of the wireless IMD sensor unit is 0%-100%RH, and the humidity detection accuracy is ±3%RH.

[0007] Further, the receiving module includes a centralized control unit and a wireless temperature receiving device, and the wireless IMD sensor unit is wirelessly connected to the centralized control unit, and the plurality of wireless temperature sensors are wirelessly connected to the wireless temperature receiving device.

[0008] A pre-warning method using the partial discharge online pre-warning device of the 35kV high-voltage switch cabinet described above, the pre-warning method comprising the following steps: S1. Multi-modal signal synchronous acquisition: TEV signal and USS signal are synchronously acquired by a wireless IMD sensor unit, wherein: The TEV signal measures the transient ground voltage signal in the range of 3-100MHz, and a contact electromagnetic induction probe is used to detect the electromagnetic pulse on the surface of the cabinet; The USS signal measures the ultrasonic signal in the range of 20-200kHz, and a piezoelectric ceramic sensor is used to capture the acoustic characteristics of partial discharge; at the same time, a high-frequency current sensor is used to acquire the high-frequency current signal in the frequency band of 10kHz-30MHz, and the sensor uses a Rogowski coil structure with a detection sensitivity not less than 1pC; S2. Multi-source signal fusion processing: The TEV signal is subjected to band-pass filtering processing to extract the effective frequency band components in the range of 3-100MHz; the USS signal is subjected to acoustic signal demodulation to separate out the characteristic frequency band in the range of 20-200kHz; The high-frequency current signal is subjected to time-frequency analysis to calculate the amplitude-phase joint characteristics of the discharge pulse; the processed three-way signals are subjected to time synchronization alignment to establish a multi-dimensional feature vector; S3. Intelligent diagnosis and pre-warning decision: a dynamic threshold judgment model is set, wherein: The TEV signal strength threshold is self-adaptively adjusted according to the environmental electromagnetic noise level; The USS signal frequency energy threshold is associated with the device vibration noise to establish a correlation function; The high-frequency current signal pulse count threshold is negatively correlated with the load current amplitude; when any two-way signal simultaneously exceeds the corresponding threshold, a primary pre-warning is triggered; when all three-way signals exceed the corresponding threshold and meet the phase synchronization condition, a high-level alarm is triggered; S4. Graded pre-warning response: The device state indicator light flashes and uploads the early warning code when the primary early warning is activated; the sound and light alarm is activated when the high-level alarm is activated The sound and light alarm is activated.

[0009] Further, the signal processing in the step S2 comprises: The wavelet packet decomposition is used for denoising the TEV signal, and the energy proportion of each sub-band is extracted as a characteristic quantity; The Mel cepstrum coefficient analysis is performed on the USS signal to construct a voiceprint feature template; The pulse shape recognition algorithm is used for the high-frequency current signal to distinguish the real discharge from the interference pulse.

[0010] Further, the dynamic threshold determination model comprises: a three-dimensional feature space clustering model established based on historical operation data; and the root mean square value and the peak-to-peak value ratio of the signal characteristics are calculated by using a sliding time window.

[0011] Further, the phase synchronization condition determination comprises: the time difference between the front edge of the TEV signal and the rising edge of the USS signal is not more than 2us; the fixed phase relationship between the high-frequency current pulse and the peak value of the TEV signal is verified; and the correlation algorithm is used to confirm the correlation of the three signals in the time domain.

[0012] Compared with the prior art, the beneficial effects of the present application are: The partial discharge online early warning device of the 35kV high-voltage switch cabinet of the present application can monitor the temperature of the switch cabinet through the temperature sensor, discover the thermal fault hidden danger in advance and make an alarm, and monitor the discharge signal of the switch cabinet, thereby improving the reliability of the safe operation of the high-voltage switch cabinet. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The embodiments of the present application will be described below.

[0014] Embodiment 1 The partial discharge online early warning device of the 35kV high-voltage switch cabinet comprises a collection module, a receiving module and a master station module. The data collected by the collection module is transmitted to the receiving module wirelessly, and the receiving module transmits the data to the master station module through an RS485 module. The collection module comprises a plurality of wireless temperature sensors, wireless IMD sensor units and high-frequency current sensors. The high-frequency current sensors are electrically connected to the wireless IMD sensor units. The wireless IMD sensor units and the high-frequency current sensors are detachably installed in the high-voltage cable compartment of the high-voltage switch cabinet.

[0015] Specifically, the 35kV high-voltage cabinet partial discharge online early warning device of the application comprises a collection module, a receiving module and a master station module. The master station module is mainly a background system, which can be used for user management and viewing of the entire system operation. The receiving module comprises a centralized control unit and a wireless temperature receiving device, which is responsible for accepting the signals transmitted by the collection module. The centralized control unit is used for receiving the data collected by the wireless IMD sensor unit, and the wireless temperature receiving device is used for receiving the collected data of the wireless temperature sensor. After data conversion and analysis, the device can not only display the data itself, but also transmit the signals to the master station module for processing through RS485 or GPRS. The master station module further comprises a display for displaying the processed data. The collection module comprises a plurality of wireless temperature sensors, wireless IMD sensor units and high-frequency current sensors. The high-frequency current sensor is electrically connected with the wireless IMD sensor unit. The wireless IMD sensor unit and the high-frequency current sensor are detachably installed in the high-voltage cable compartment of the high-voltage switch cabinet. The wireless IMD sensor has an ultrasonic sensor, a TEV sensor and a humidity sensor built-in, and can externally connect an HFCT sensor through a coaxial cable to collect discharge signals at multiple angles. The collection module receives the collected signals of the temperature sensor in a wireless manner, and then transmits all the signals to the master station module for processing in a wireless transmission manner.

[0016] The collection module and the receiving module of the early warning device communicate in a wireless manner, which can automatically match the address and avoid the trouble of on-site wiring caused by wired mode. The early warning device can effectively monitor the internal air gap, crack fault, insulation aging, discharge signal on the surface of the insulation material and the change of the humidity in the switch cabinet caused by the failure of the main insulation material in the switch cabinet. It can also monitor the temperature overload fault caused by the poor contact or overload of the circuit breaker and the connection line in the switch cabinet through a plurality of wireless temperature sensors. The passive wireless temperature sensor communicates in a wireless manner and is self-powered without the need for a power supply. The high-frequency current sensor has two types of open type and closed type, which can be selected according to the site conditions.

[0017] In some embodiments, the measurement range of the USS measurement of the wireless IMD sensor unit is 0-60dB, the resolution is 1dB, the accuracy is ±1dB, and the center frequency is 40KHz. The linearity error of the TEV measurement of the wireless IMD sensor unit is within ±10%, and the working frequency band is 3-100MHz. The humidity detection range of the humidity measurement of the wireless IMD sensor unit is 0%-100%RH, and the humidity detection accuracy is ±3%RH.

[0018] Specifically, the wireless IMD sensor unit adopts a TEV, ultrasonic sensor, and humidity sensor three-in-one integrated design, can be externally connected to an HFCT sensor, the wireless sensor unit is installed in the high-voltage cable compartment of the high-voltage switch cabinet, and is fixed by screws and installed by magnetic adsorption. A 2P air switch is additionally installed in the switch cabinet, and the power supply is output from the lighting and 220V alternating current according to the site conditions. The specific installation method is as follows: for newly built substations, the screw fixing method is adopted, and for transformed substations, the magnetic adsorption installation method is adopted.

[0019] In some embodiments, the receiving module includes a centralized control unit and a wireless temperature receiving device, the wireless IMD sensor unit is wirelessly connected to the centralized control unit, and the wireless temperature sensors are wirelessly connected to the wireless temperature receiving device.

[0020] Specifically, according to the site conditions, the wireless temperature receiving device can be fixed on the wall by screws, and a power line is separately taken 220V power supply, and the wireless temperature sensors are connected wirelessly; the wireless temperature sensor is a passive wireless temperature sensor, which can be installed by binding on the busbar, circuit breaker, upper and lower contacts, or inlet and outlet terminals, and the module is wound around the power supply part (main current) to make the temperature measuring point contact the part to be measured.

[0021] Embodiment 2 A pre-warning method of the partial discharge on-line pre-warning device of the 35kV high-voltage switch cabinet described above, the pre-warning method includes the following steps: S1. Multi-modal signal synchronous acquisition: TEV signal and USS signal are synchronously acquired by the wireless IMD sensor unit, wherein: The TEV signal measures the transient ground voltage signal in the range of 3-100MHz, and a contact electromagnetic induction probe is used to detect the electromagnetic pulse on the surface of the cabinet; The USS signal measures the ultrasonic signal in the range of 20-200kHz, and a piezoelectric ceramic sensor is used to capture the acoustic characteristics of partial discharge; at the same time, a high-frequency current sensor is used to acquire the high-frequency current signal in the frequency band of 10kHz-30MHz, and the sensor adopts a Rogowski coil structure with a detection sensitivity not less than 1pC; S2. Multi-source signal fusion processing: The TEV signal is subjected to band-pass filtering processing to extract the effective frequency band component in the range of 3-100MHz; the USS signal is subjected to acoustic signal demodulation to separate out the characteristic frequency band in the range of 20-200kHz; The high-frequency current signal is subjected to time-frequency analysis to calculate the amplitude-phase joint characteristics of the discharge pulse; the processed three-way signals are time-synchronously aligned to establish a multi-dimensional feature vector; S3. Intelligent diagnosis and pre-warning decision: a dynamic threshold judgment model is set, wherein: TEV signal intensity threshold is self-adaptively adjusted according to the environmental electromagnetic noise level; USS signal frequency domain energy threshold is associated with the device vibration noise; High-frequency current signal pulse count threshold is negatively correlated with the load current amplitude; when any two signals simultaneously exceed the corresponding threshold, a primary warning is triggered; when all three signals exceed the corresponding threshold and meet the phase synchronization condition, a high-level alarm is triggered; S4. Hierarchical warning response: The device status indicator light flashes and uploads the warning code when the primary warning is triggered; when the high-level alarm is triggered The audible and visual alarms are activated.

[0022] Multi-source signal fusion processing: TEV signal processing: Band-pass filtering: an 8th-order Chebyshev filter is used, with a passband of 3-100 MHz and a stopband attenuation of ≥60 dB.

[0023] Wavelet packet decomposition denoising: db4 wavelet basis is used for 4-layer decomposition, and the energy proportion of each sub-band (such as 30-50 MHz, 50-80 MHz) is extracted, with the calculation formula being: (i=1,2,...,16) where W i (t) is the wavelet packet coefficient of the i-th sub-band.

[0024] USS signal processing: Acoustic wave demodulation: envelope detection (diode peak detection circuit) is performed on the original signal to extract the 20-200 kHz modulation component.

[0025] Mel frequency cepstral coefficient (MFCC) analysis: frame and windowing (frame length 256 points, Hamming window), power spectrum calculation. Through a 40-channel Mel filter bank, log energy features are extracted. After discrete cosine transform (DCT), the first 12-dimensional coefficients are taken to construct a voiceprint template library.

[0026] HFCT signal processing: Time-frequency analysis: short-time Fourier transform (STFT) is used, with a window length of 1024 points and a step length of 512 points, to extract the amplitude-phase joint features of the discharge pulse, such as: Pulse rise time (10%-90% level): tr<10nstr<10ns Phase difference: adjacent pulse peak phase difference ≤5° Feature fusion: The feature vectors (TEV energy ratio, MFCC coefficient, and HFCT pulse parameters) of the three signals are concatenated according to the time window (100ms) to form a multidimensional feature vector (dimension: 16+12+3=31 dimensions).

[0027] Dynamic threshold model construction: Three-dimensional feature space clustering: A Gaussian mixture model (GMM) is established based on historical data (normal / discharge state) to divide the feature space into normal area, transition area, and discharge area.

[0028] Threshold adaptive adjustment: TEV threshold: V th =μ noise +3σ nois , where μ noise is the mean value of the ambient electromagnetic noise.

[0029] USS energy threshold: E th =kf(v), f(v) is the vibration velocity of the equipment, which can be measured by an acceleration sensor, k=0.1-0.3.

[0030] HFCT pulse counting threshold: N th =10−0.05I load , I load is the load current (unit A) Phase synchronization determination: Calculate the time difference between the leading edge of the TEV signal and the rising edge of the USS signal: Δt=arg max (R TEV,USS (τ)) requires Δt≤2μs, where R TEV，USS为互相关函数。 Phase relationship between HFCT pulse and TEV signal: Verify that the HFCT pulse peak appears within 50-200ns after the rising edge of the TEV signal.

[0031] The technical solutions of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that the above descriptions are merely for the purpose of explaining the solutions of the present invention and are not to be construed in any way as limiting the scope of protection of the invention. Based on the explanations herein, those skilled in the art can conceive of other specific embodiments of the present invention or equivalent replacements without inventive effort, and these will fall within the scope of protection of the present invention.

Claims

1. A partial discharge online early warning device for a 35kV high-voltage switchgear, characterized in that: It includes an acquisition module, a receiving module and a master station module. The data collected by the acquisition module is wirelessly transmitted to the receiving module, and the receiving module transfers the data to the master station module through the RS485 module. The acquisition module includes several wireless temperature sensors, wireless IMD sensor units and high-frequency current sensors. The high-frequency current sensor is electrically connected to the wireless IMD sensor unit. The wireless IMD sensor unit and the high-frequency current sensor are detachably installed in the high-voltage cable compartment of the high-voltage switch cabinet.

2. The 35kV high-voltage switchgear partial discharge online early warning device according to claim 1 is characterized in that: The USS measurement of the wireless IMD sensor unit has a measurement range of 0-60dB, a resolution of 1dB, an accuracy of ±1dB, and a center frequency of 40KHz. The linearity error of the TEV measurement of the wireless IMD sensor unit is within ±10%, and the operating frequency band is 3-100MHz. The humidity detection range of the humidity measurement of the wireless IMD sensor unit is 0%-100%RH, and the humidity detection accuracy is ±3%RH.

3. The online early warning device for partial discharge of a 35kV high-voltage switchgear according to claim 1 is characterized in that: The receiving module includes a central control unit and a wireless temperature receiving device. The wireless IMD sensor unit is wirelessly connected to the central control unit, and a plurality of the wireless temperature sensors are wirelessly connected to the wireless temperature receiving device.

4. An early warning method using the online early warning device for partial discharge of a 35kV high-voltage switchgear according to any one of claims 1 to 3, characterized in that: The early warning method comprises the following steps: S1. Synchronous acquisition of multimodal signals: Synchronous acquisition of TEV and USS signals via wireless IMD sensor units, including: The TEV signal measurement range is 3-100MHz transient ground voltage signal, and a contact electromagnetic induction probe is used to detect electromagnetic pulses on the cabinet surface; The USS signal measurement range is 20-200kHz, using piezoelectric ceramic sensors to capture the acoustic characteristics of partial discharges. High-frequency current sensors in the 10kHz-30MHz band are also used to collect high-frequency current signals. These sensors use a Rogowski coil structure and have a detection sensitivity of no less than 1pC. S2. Multi-source signal fusion processing: Perform bandpass filtering on the TEV signal to extract the 3-100 MHz effective frequency band component; perform acoustic signal demodulation on the USS signal to separate the 20-200 kHz characteristic frequency band; Performing time-frequency analysis on the high-frequency current signal to calculate the amplitude-phase joint characteristics of the discharge pulse; performing time synchronization alignment on the processed three signals to establish a multi-dimensional feature vector; S3. Intelligent diagnosis and early warning decision-making: Set up a dynamic threshold judgment model, where: The TEV signal strength threshold is adaptively adjusted according to the environmental electromagnetic noise level; Establish a correlation function between the frequency domain energy threshold of the USS signal and the equipment vibration noise; The high-frequency current signal pulse count threshold is negatively correlated with the load current amplitude; when any two signals exceed the corresponding threshold at the same time, a primary warning is triggered; when all three signals exceed the corresponding threshold and meet the phase synchronization conditions, a high-level warning is triggered. police; S4. Graded warning response: When the primary warning is triggered, the device status indicator light flashes and the warning code is uploaded; when the high-level alarm is triggered, the device status indicator light flashes and the warning code is uploaded. Activate the audible and visual alarms.

5. The early warning method according to claim 4, characterized in that: The signal processing in step S2 includes: Wavelet packet decomposition is used to denoise the TEV signal, and the energy proportion of each sub-band is extracted as the feature value; Perform Mel cepstral coefficient analysis on USS signals to construct voiceprint feature templates; A pulse shape recognition algorithm is used on high-frequency current signals to distinguish real discharges from interference pulses.

6. The early warning method according to claim 4, characterized in that: The dynamic threshold determination model includes: a three-dimensional feature space clustering model established based on historical operation data; and a sliding time window for calculating the root mean square value and peak-to-peak ratio of signal features.

7. The early warning method according to claim 4, characterized in that: The phase synchronization condition determination includes: calculating that the time difference between the leading edge of the TEV signal and the rising edge of the USS signal does not exceed 2μs; verifying that a fixed phase relationship exists between the high-frequency current pulse and the peak of the TEV signal; and using a cross-correlation algorithm to confirm the correlation of the three signals in the time domain.