High-voltage power equipment partial discharge hidden fault monitoring system and method

Abstract

The invention belongs to the technical field of high-voltage power equipment monitoring, and particularly relates to a high-voltage power equipment partial discharge hidden fault monitoring system and method, and the system comprises a multi-sensor fusion sensing module, an anti-interference signal processing module, an intelligent diagnosis positioning module, and a dynamic threshold early warning module. The multi-sensing fusion sensing module adopts an ultrahigh frequency, ultrasonic wave and high-frequency current three-sensing combined framework to realize signal complementary acquisition; the anti-interference signal processing module purifies signals in a mode of combining hardware shielding and algorithm filtering; the intelligent diagnosis and positioning module realizes accurate diagnosis and positioning based on multi-dimensional feature extraction, fault fingerprint database comparison and multi-source fusion positioning; and the dynamic threshold value early warning module dynamically adjusts an early warning threshold value in combination with equipment working conditions and performs graded early warning. According to the technical scheme, the problems that in the prior art, partial discharge signals are difficult to capture, the fault positioning precision is low, and the false alarm and missing alarm rate is high can be solved, and safe and stable operation of high-voltage power equipment is effectively guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of high-voltage power equipment monitoring technology, specifically relating to a system and method for monitoring hidden partial discharge faults in high-voltage power equipment. Background Technology High-voltage power equipment (such as transformers, GIS switchgear, and cable terminals) is the core carrier for the safe and stable operation of power systems, and its insulation performance directly determines the reliability of power grid operation. Partial discharge is a typical early characteristic of insulation degradation in high-voltage equipment. It refers to the non-penetrating discharge phenomenon that occurs in a local area of ​​the insulating medium inside the equipment under the action of an electric field. The discharge amount is usually only 10-1000 pC, making it highly concealed. As power systems develop towards higher voltage, larger capacity, and intelligence, the operating conditions of equipment are becoming increasingly complex, and the risk of insulation faults caused by partial discharge has significantly increased. According to statistics on equipment faults from the State Grid Corporation of China, approximately 65% ​​of major faults in high-voltage equipment originate from the gradual degradation of insulation caused by partial discharge. However, due to technological limitations, traditional detection methods have a capture rate of less than 30% for early, weak partial discharge signals. Often, by the time the fault becomes apparent, equipment damage and power outages have already occurred, resulting in huge economic losses and social impacts.

[0002] Traditional offline detection methods (such as periodic offline withstand voltage tests and manual inspections) have significant shortcomings: first, the detection cycle is long (usually six months to one year), making real-time monitoring difficult; second, they are greatly affected by electromagnetic interference and have poor ability to identify weak discharge signals below 100pC; and third, they require power outages during testing, affecting the reliability of the power grid. Existing monitoring technologies also suffer from difficulties in capturing weak signals, insufficient fault location accuracy, high false alarm and false negative rates, and insufficient real-time and continuous operation, failing to meet the urgent needs for safe operation of high-voltage power equipment. Summary of the Invention

[0003] This invention provides a partial discharge hidden fault monitoring system and method for high-voltage power equipment, which solves the problems of difficulty in capturing partial discharge signals, low fault location accuracy, high false alarm and false alarm rates, and insufficient real-time continuity in the prior art.

[0004] It includes a multi-sensor fusion sensing module, an anti-interference signal processing module, an intelligent diagnosis and positioning module, and a dynamic threshold early warning module, with each module connected in sequence. The multi-sensor fusion sensing module adopts a three-sensor joint monitoring architecture of "UHF + ultrasound (AE) + high frequency current (HFCT)" to achieve complementary acquisition of partial discharge signals. The anti-interference signal processing module adopts a dual anti-interference strategy of "hardware shielding + algorithm filtering" to purify the acquired raw signal. The intelligent diagnostic and positioning module is used to achieve intelligent identification of partial discharge faults and precise positioning of discharge points. The dynamic threshold early warning module is used to dynamically adjust the early warning threshold according to the equipment operating conditions and to provide tiered early warnings.

[0005] Based on the above technical solution, the multi-sensor fusion sensing module further includes: Ultra-high frequency sensors are installed inside or on the outer wall of the equipment enclosure to collect 300MHz-3GHz ultra-high frequency electromagnetic waves generated by partial discharge. The detection rate of weak discharge signals at the 10pC level is over 90%. Ultrasonic sensors are uniformly arranged against the equipment shell with a spacing of ≤1.5m, and are used to collect ultrasonic signals of 20kHz-200kHz generated by discharge. A high-frequency current sensor is fitted onto the equipment's grounding wire or iron core grounding wire to collect 1MHz-100MHz high-frequency pulse current generated by discharge, serving as an interference identification benchmark.

[0006] Based on the above technical solution, the anti-interference signal processing module further includes: The hardware anti-interference unit includes a sensor encapsulated in a metal shielded shell, a twisted-pair shielded transmission cable, and a data acquisition unit with a built-in power frequency notch filter. The power frequency notch filter is used to filter out 50Hz fundamental and harmonic interference. The anti-interference algorithm unit adopts a combination algorithm of "wavelet transform + adaptive noise cancellation". Wavelet transform is used to decompose the signal into different frequency bands and extract the characteristic frequency band of the discharge signal. Adaptive noise cancellation uses the interference signal collected by the high-frequency current sensor as a reference to cancel external interference.

[0007] Based on the above technical solution, the intelligent diagnostic positioning module further includes: The feature extraction unit is used to extract multi-dimensional feature parameters such as discharge phase distribution, pulse amplitude spectrum, pulse width, and repetition frequency from the processed signal to construct a partial discharge feature vector. The fault fingerprint database stores feature vectors collected through laboratory simulations of different fault types, with a sample size of ≥1000. The fault types include insulation gap discharge, surface discharge, metal tip discharge, and insulation aging discharge. The intelligent diagnostic unit uses an improved random forest algorithm to compare the feature vectors collected on-site with the fault fingerprint database, achieving a fault type identification accuracy of over 92%. The multi-source fusion positioning unit, based on the time difference positioning method of ultrasonic sensor array, combines the signal intensity distribution of UHF sensor and fuses the two positioning results through particle filtering algorithm to achieve three-dimensional positioning of the discharge point with a positioning error ≤10cm.

[0008] Based on the above technical solution, the dynamic threshold early warning module further includes: The basic threshold setting unit sets the initial discharge threshold according to the equipment type and rated voltage. The initial threshold for 110kV transformers is 50pC, and the initial threshold for 220kV GIS is 30pC. The dynamic adjustment unit incorporates three adjustment factors: equipment operating years (aging factor), real-time load rate (load factor), and ambient temperature (temperature factor), and dynamically adjusts the threshold through a multiple linear regression model. The graded early warning unit is set with level 1, level 2 and level 3 early warning. Level 1 early warning is when the discharge quantity reaches 80% of the dynamic threshold and the duration is >1 hour. Level 2 early warning is when the discharge quantity reaches the dynamic threshold and the feature matches the fault fingerprint. Level 3 early warning is when the discharge quantity exceeds 150% of the dynamic threshold or the discharge is located in a critical insulation part, and a diagnostic report and maintenance suggestions are pushed.

[0009] This invention also provides a method for monitoring hidden partial discharge faults in high-voltage power equipment, comprising the following steps: S1: The partial discharge signal of the high-voltage power equipment is collected through a multi-sensor fusion sensing module. The three-sensor joint acquisition method of ultra-high frequency, ultrasonic and high frequency current is adopted to achieve signal complementarity. S2: The original signal is purified by the anti-interference signal processing module, and a dual anti-interference strategy combining hardware shielding and algorithm filtering is adopted. S3: The purified signal is processed by the intelligent diagnostic and positioning module, multi-dimensional feature parameters are extracted to construct feature vectors, and the fault type is identified by comparing with the fault fingerprint database. Combined with the multi-source positioning algorithm, the discharge point is accurately located. S4: The dynamic threshold warning module dynamically adjusts the warning threshold according to the equipment operating conditions to provide graded warnings for partial discharge faults.

[0010] Furthermore, in step S1, the ultra-high frequency sensor collects ultra-high frequency electromagnetic waves of 300MHz-3GHz, the ultrasonic sensor collects ultrasonic signals of 20kHz-200kHz, and the high-frequency current sensor collects high-frequency pulse current of 1MHz-100MHz.

[0011] Furthermore, in step S2, the hardware shielding includes sensor metal shielding encapsulation and twisted-pair shielded transmission cable transmission. The algorithm filtering uses wavelet transform to extract the characteristic frequency band of the discharge signal and combines adaptive noise cancellation to cancel external interference, thereby improving the signal-to-noise ratio of the discharge signal by 10-100 times.

[0012] Furthermore, in step S3, multi-dimensional feature parameters are extracted to construct feature vectors, and an improved random forest algorithm is used for fault identification. A multi-source fusion positioning algorithm combining ultrasonic sensor time difference positioning method and UHF sensor signal intensity distribution is used to achieve three-dimensional positioning of the discharge point with a positioning error ≤10cm.

[0013] Furthermore, in step S4, the early warning threshold is dynamically adjusted based on the equipment's operating years, real-time load rate, and ambient temperature, and diagnostic reports and maintenance suggestions are pushed out according to the first-level, second-level, and third-level early warning mechanisms.

[0014] This invention employs an integrated technical approach of "multi-sensor fusion sensing + anti-interference signal processing + intelligent diagnostic positioning + dynamic threshold early warning" to construct a real-time, accurate, and anti-interference partial discharge hidden fault monitoring system. The multi-sensor fusion sensing module achieves complementary signal acquisition through a three-sensor joint architecture, enhancing the ability to capture weak signals; the anti-interference signal processing module adopts a dual anti-interference strategy combining hardware and algorithms to effectively purify signals; the intelligent diagnostic positioning module achieves accurate fault identification and location based on multi-dimensional feature extraction and multi-source fusion positioning; and the dynamic threshold early warning module dynamically adjusts thresholds according to equipment operating conditions, reducing false alarm and missed alarm rates.

[0015] The beneficial effects of this invention are: Strong ability to capture weak signals: The detection rate of weak discharge signals at the 10pC level is over 90%, and it can effectively capture intermittent and transient partial discharge signals; High fault location accuracy: It achieves three-dimensional positioning of the discharge point with an error of ≤10cm, which greatly improves maintenance efficiency; Low false alarm and false alarm rates: False alarm rate is less than 5%, and false alarm rate is less than 2%, effectively distinguishing partial discharge from external interference; Good real-time continuity: It supports 24-hour continuous monitoring of equipment under load without power outages, and can reflect the discharge situation under actual operating conditions; Timely and accurate early warning: partial discharge hidden faults can be detected 3-10 months in advance, allowing sufficient time for maintenance, greatly reducing the failure rate and maintenance costs of equipment, and significantly shortening power outage time.

[0016] Other features and beneficial effects of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects of the invention and other beneficial effects may be realized and obtained by means of the structures particularly pointed out in the description and claims. Attached Figure Description

[0017] Figure 1 This is a structural diagram of a partial discharge hidden fault monitoring system for high-voltage power equipment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. The technical features designed in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] In the description of this invention, it should be noted that all terms used in this invention (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains, and should not be construed as limiting the invention; it should be further understood that the terms used in this invention should be understood to have the same meaning as those in the context of this specification and in the relevant field, and should not be understood in an idealized or overly formal sense, except as expressly defined in this invention.

[0020] The present invention provides the following embodiments: like Figure 1 As shown, a partial discharge hidden fault monitoring system for high-voltage power equipment includes a multi-sensor fusion sensing module, an anti-interference signal processing module, an intelligent diagnostic positioning module, and a dynamic threshold early warning module, with each module connected in sequence.

[0021] The multi-sensor fusion sensing module adopts a three-sensor joint monitoring architecture of "UHF + ultrasound (AE) + high-frequency current (HFCT)" to achieve complementary signal acquisition and improve the ability to capture weak signals. Among them, the UHF sensor is installed inside or on the outer wall of the equipment enclosure to collect 300MHz-3GHz UHF electromagnetic waves generated by partial discharge. This frequency band avoids power frequency electromagnetic interference (50Hz), has low signal attenuation, can penetrate insulating media, and has a detection rate of over 90% for weak discharge signals at the 10pC level. The ultrasound sensors are evenly arranged against the equipment shell with a spacing of ≤1.5m to collect 20kHz-200kHz ultrasonic signals generated by discharge. The ultrasonic signal is not affected by electromagnetic interference, and spatial positioning of the discharge point can be achieved through the multi-sensor time difference positioning method. The high-frequency current sensor is fitted onto the equipment grounding wire or iron core grounding wire to collect 1MHz-100MHz high-frequency pulse current generated by discharge. This signal can be used as an "interference identification benchmark" to distinguish between internal discharge and external interference.

[0022] The anti-interference signal processing module employs a dual anti-interference strategy of "hardware shielding + algorithm filtering" to achieve signal purification. In the hardware anti-interference unit, the sensor is encapsulated in a metal shielded shell, and the transmission line uses twisted-pair shielded cable to reduce electromagnetic interference coupling; the data acquisition unit has a built-in power frequency notch filter to filter out 50Hz fundamental and harmonic interference. The algorithm anti-interference unit uses a combination algorithm of "wavelet transform + adaptive noise cancellation" to process the acquired raw signal. Wavelet transform decomposes the signal into different frequency bands, extracting the characteristic frequency band of the discharge signal (such as the 500MHz-2GHz band of UHF signals); adaptive noise cancellation uses the interference signal acquired by the HFCT as a reference to cancel external interference, improving the signal-to-noise ratio of the discharge signal by 10-100 times.

[0023] The intelligent diagnostic and positioning module constructs an intelligent diagnostic system based on "feature extraction + fingerprint comparison + multi-source fusion positioning". The feature extraction unit extracts multi-dimensional feature parameters such as discharge phase distribution, pulse amplitude spectrum, pulse width, and repetition frequency from the processed signal to construct a partial discharge feature vector. The fault fingerprint database collects feature vectors for various fault types (such as insulation gap discharge, surface discharge, metal tip discharge, and insulation aging discharge) through laboratory simulations, establishing a fault fingerprint database containing over 1000 samples, including different equipment types. The intelligent diagnostic unit uses an improved random forest algorithm to compare the feature vectors collected on-site with the fingerprint database, achieving a fault type identification accuracy of over 92%. The multi-source fusion positioning unit, based on the time difference positioning method of ultrasonic sensor arrays and combined with the signal intensity distribution of UHF sensors, fuses the two positioning results through a particle filtering algorithm to achieve three-dimensional positioning of the discharge point with an error ≤10cm.

[0024] The dynamic threshold early warning module establishes a dynamic threshold model that is "adaptive to operating conditions and covers the entire life cycle". The basic threshold setting unit sets the initial discharge threshold based on the equipment type (such as transformer, GIS) and rated voltage (e.g., the initial threshold for a 110kV transformer is 50pC, and the initial threshold for a 220kV GIS is 30pC). The dynamic adjustment unit introduces three adjustment factors: equipment operating years (aging factor), real-time load rate (load factor), and ambient temperature (temperature factor). It dynamically adjusts the threshold through a multiple linear regression model. For example, the aging factor of a 10-year-old transformer is 0.6, and the load factor is 1.2 when the load rate is 80%, so the threshold is adjusted to 50 × 0.6 × 1.2 = 36pC. The graded early warning unit sets a first-level warning (discharge reaches 80% of the dynamic threshold and the duration is >1 hour), a second-level warning (discharge reaches the dynamic threshold and the feature matches the fault fingerprint), and a third-level warning (discharge exceeds 150% of the dynamic threshold, or the discharge is located in a critical insulation part), and pushes a diagnostic report and maintenance suggestions.

[0025] The method for monitoring hidden partial discharge faults in high-voltage power equipment according to the present invention includes the following steps: S1: The partial discharge signal of high-voltage power equipment is collected by the multi-sensor fusion sensing module, the ultra-high frequency sensor collects 300MHz-3GHz ultra-high frequency electromagnetic waves, the ultrasonic sensor collects 20kHz-200kHz ultrasonic signals, and the high-frequency current sensor collects 1MHz-100MHz high-frequency pulse current to achieve signal complementarity. S2: The original signal collected is purified by the anti-interference signal processing module. The hardware shielding includes metal shielding of the sensor and transmission of twisted pair shielded transmission cable. The algorithm filtering uses wavelet transform to extract the characteristic frequency band of the discharge signal and combines adaptive noise cancellation to cancel external interference, thereby improving the signal-to-noise ratio of the discharge signal by 10-100 times. S3: The purified signal is processed by the intelligent diagnostic positioning module, multi-dimensional feature parameters are extracted to construct feature vectors, and the fault type is identified by comparing the improved random forest algorithm with the fault fingerprint database. The three-dimensional positioning of the discharge point is achieved by combining the ultrasonic sensor time difference positioning method with the signal intensity distribution of the UHF sensor. The positioning error is ≤10cm. S4: The dynamic threshold early warning module dynamically adjusts the early warning threshold based on the equipment's operating years, real-time load rate, and ambient temperature, and pushes diagnostic reports and maintenance suggestions according to the first-level, second-level, and third-level early warning mechanisms.

[0026] The effectiveness of the present invention is verified through specific embodiments below: Example 1: Partial Discharge Monitoring of a 110kV Oil-Immersed Transformer Test conditions: A 110kV oil-immersed transformer in a substation (8 years of operation, rated capacity 50MVA) with a preset insulation gap discharge fault (discharge amount of about 50pC, intermittent discharge, cycle of about 30 minutes).

[0027] The monitoring system of this invention is used to install two UHF sensors on the top of the transformer tank, four ultrasonic sensors are evenly arranged on the side wall, and one HFCT sensor is installed on the grounding wire for continuous 24-hour monitoring.

[0028] Results: The system captured the first discharge signal within 1 hour of operation and completed the fault type identification (insulation air gap discharge) within 3 hours. The discharge point was located at the beginning of the high-voltage winding (the actual position and the positioning result had an error of 6.5cm). During the 1-year test period, a total of 12 warnings were issued, with no missed or false alarms. The maintenance personnel accurately repaired the fault based on the positioning results, and the equipment returned to normal operation.

[0029] Example 2: Partial Discharge Monitoring of 220kV GIS Switchgear Test conditions: A 220kV GIS switchgear in a converter station (5 years of operation, rated current 2000A) has an internal metal tip discharge fault (discharge amount of about 30pC, continuous discharge), and the operating environment is subject to strong electromagnetic interference (there is a 500kV busbar nearby, and the electromagnetic field strength is about 20kV / m).

[0030] The monitoring system of this invention uses UHF sensors installed at the basin-type insulators of GIS switchgear, three ultrasonic sensors arranged on the outer casing, and an HFCT sensor installed on the grounding wire, with anti-interference mode enabled.

[0031] Results: The system effectively canceled electromagnetic interference through hardware shielding and algorithm filtering, improving the signal-to-noise ratio of the discharge signal from 0.5 to 15; within 2 hours of operation, it identified a metal tip discharge fault and located the discharge point at the bus joint (error 7.8cm); after issuing a level-two warning, maintenance personnel completed the fault handling without power interruption, and the equipment returned to normal without causing any power outage losses.

[0032] Example 3: Partial Discharge Monitoring of 10kV Cable Terminals Test conditions: A 10kV cable terminal in an industrial park (3 years of operation, load rate fluctuation range of 40%-90%) has an insulation aging discharge fault (discharge amount of about 80pC, intermittent discharge with load changes).

[0033] The monitoring system of this invention uses a UHF sensor and an ultrasonic sensor installed at the cable terminal head, an HFCT sensor installed on the grounding wire, and a dynamic threshold module enabled (initial threshold 100pC, dynamically adjusted according to the load rate).

[0034] Results: The system dynamically adjusts the threshold according to the load rate change (the threshold drops to 70pC when the load rate is 90%), captures the discharge signal during the peak load, identifies the insulation aging fault within 3 hours, and locates the discharge point at the terminal insulation skirt (error 6.0cm); after issuing the first-level warning, the maintenance personnel arrange for power outage maintenance, replace the cable terminal, and prevent the fault from expanding.

[0035] The sensors and hardware units used in the technical solution of this invention are commercially available products and are existing technologies. Furthermore, other conventional technologies are not specifically described. The results of the above embodiments demonstrate that the high-voltage power equipment partial discharge concealed fault monitoring system and method of this invention exhibits excellent performance in key indicators such as weak signal capture, anti-interference capability, fault identification and location accuracy, and early warning accuracy, possessing extremely high engineering feasibility and promotional value.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A partial discharge concealed fault monitoring system for high-voltage power equipment, characterized in that, It includes a multi-sensor fusion sensing module, an anti-interference signal processing module, an intelligent diagnosis and positioning module, and a dynamic threshold early warning module, with each module connected in sequence. The multi-sensor fusion sensing module includes a three-sensor joint monitoring architecture for complementary acquisition of partial discharge signals. The three-sensor joint monitoring architecture includes an ultra-high frequency sensor, an ultrasonic sensor, and a high-frequency current sensor. The anti-interference signal processing module is used to purify the acquired raw signal; The intelligent diagnostic and positioning module is used to achieve intelligent identification of partial discharge faults and precise positioning of discharge points. The dynamic threshold early warning module is used to dynamically adjust the early warning threshold according to the equipment operating conditions and to provide tiered early warnings.

2. The high-voltage power equipment partial discharge concealed fault monitoring system according to claim 1, characterized in that, The multi-sensor fusion sensing module includes: Ultra-high frequency sensors are installed inside or on the outer wall of the equipment enclosure to collect 300MHz-3GHz ultra-high frequency electromagnetic waves generated by partial discharge. The detection rate of weak discharge signals at the 10pC level is over 90%. Ultrasonic sensors are uniformly arranged against the equipment shell with a spacing of ≤1.5m, and are used to collect ultrasonic signals of 20kHz-200kHz generated by discharge. A high-frequency current sensor is fitted onto the equipment's grounding wire or iron core grounding wire to collect 1MHz-100MHz high-frequency pulse current generated by discharge, serving as an interference identification benchmark.

3. The high-voltage power equipment partial discharge concealed fault monitoring system according to claim 1, characterized in that, The anti-interference signal processing module includes: The hardware anti-interference unit includes a sensor encapsulated in a metal shielded shell, a twisted-pair shielded transmission cable, and a data acquisition unit with a built-in power frequency notch filter. The power frequency notch filter is used to filter out 50Hz fundamental and harmonic interference. The anti-interference algorithm unit adopts a combination algorithm of wavelet transform and adaptive noise cancellation. The wavelet transform is used to decompose the signal into different frequency bands and extract the characteristic frequency band of the discharge signal. The adaptive noise cancellation uses the interference signal collected by the high-frequency current sensor as a reference to cancel external interference.

4. The high-voltage power equipment partial discharge concealed fault monitoring system according to claim 1, characterized in that, The intelligent diagnostic and positioning module includes: The feature extraction unit is used to extract eight feature parameters from the processed signal, namely, discharge phase distribution, pulse amplitude spectrum, pulse width, and repetition frequency, and to construct a partial discharge feature vector. The fault fingerprint database stores feature vectors collected through laboratory simulations of different fault types, with a sample size of ≥1000. The fault types include insulation gap discharge, surface discharge, metal tip discharge, and insulation aging discharge. The intelligent diagnostic unit uses an improved random forest algorithm to compare the feature vectors collected on-site with the fault fingerprint database, achieving a fault type identification accuracy of over 92%. The multi-source fusion positioning unit, based on the time difference positioning method of ultrasonic sensor array, combines the signal intensity distribution of UHF sensor and fuses the two positioning results through particle filtering algorithm to achieve three-dimensional positioning of the discharge point with a positioning error ≤10cm.

5. The high-voltage power equipment partial discharge concealed fault monitoring system according to claim 1, characterized in that, The dynamic threshold early warning module includes: The basic threshold setting unit sets the initial discharge threshold according to the equipment type and rated voltage. The initial threshold for 110kV transformers is 50pC, and the initial threshold for 220kV GIS is 30pC. The dynamic adjustment unit incorporates three adjustment factors: equipment operating years, real-time load rate, and ambient temperature, and dynamically adjusts the threshold through a multiple linear regression model. The graded early warning unit is set with level 1, level 2 and level 3 early warning. Level 1 early warning is when the discharge quantity reaches 80% of the dynamic threshold and the duration is >1 hour. Level 2 early warning is when the discharge quantity reaches the dynamic threshold and the feature matches the fault fingerprint. Level 3 early warning is when the discharge quantity exceeds 150% of the dynamic threshold or the discharge is located in a critical insulation part, and a diagnostic report and maintenance suggestions are pushed.

6. A method for monitoring hidden partial discharge faults in high-voltage power equipment, characterized in that, Includes the following steps: S1: The partial discharge signal of the high-voltage power equipment is collected through a multi-sensor fusion sensing module. The three-sensor joint acquisition method of ultra-high frequency, ultrasonic and high frequency current is adopted to achieve signal complementarity. S2: The original signal is purified by the anti-interference signal processing module, and a dual anti-interference strategy combining hardware shielding and algorithm filtering is adopted. S3: The purified signal is processed by the intelligent diagnostic and positioning module, multi-dimensional feature parameters are extracted to construct feature vectors, and the fault type is identified by comparing with the fault fingerprint database. Combined with the multi-source positioning algorithm, the discharge point is accurately located. S4: The dynamic threshold warning module dynamically adjusts the warning threshold according to the equipment operating conditions to provide graded warnings for partial discharge faults.

7. The method for monitoring hidden partial discharge faults in high-voltage power equipment according to claim 6, characterized in that, In step S1, the ultra-high frequency sensor collects ultra-high frequency electromagnetic waves of 300MHz-3GHz, the ultrasonic sensor collects ultrasonic signals of 20kHz-200kHz, and the high frequency current sensor collects high frequency pulse current of 1MHz-100MHz.

8. The method for monitoring hidden partial discharge faults in high-voltage power equipment according to claim 6, characterized in that, In step S2, the hardware shielding includes sensor metal shielding encapsulation and twisted pair shielded transmission cable transmission. The algorithm filtering uses wavelet transform to extract the characteristic frequency band of the discharge signal and combines adaptive noise cancellation to cancel external interference, thereby improving the signal-to-noise ratio of the discharge signal by 10-100 times.

9. The method for monitoring hidden partial discharge faults in high-voltage power equipment according to claim 6, characterized in that, In step S3, multi-dimensional feature parameters are extracted to construct feature vectors, and an improved random forest algorithm is used for fault identification. A multi-source fusion positioning algorithm combining ultrasonic sensor time difference positioning method and UHF sensor signal intensity distribution is used to achieve three-dimensional positioning of the discharge point with a positioning error ≤10cm.

10. The method for monitoring hidden partial discharge faults in high-voltage power equipment according to claim 6, characterized in that, In step S4, the early warning threshold is dynamically adjusted based on the equipment's operating years, real-time load rate, and ambient temperature, and diagnostic reports and maintenance suggestions are pushed out according to the first-level, second-level, and third-level early warning mechanisms.

Images