Stator winding end portion corona discharge detection method based on multi-signal joint monitoring

Abstract

The invention discloses a stator winding end portion corona discharge detection method based on multi-signal joint monitoring, and relates to the technical field of large generator stator winding end portion corona discharge detection, and the method comprises the steps: S1, collecting corona discharge signals; s2, processing a corona discharge signal; s3, judging the state of corona discharge; and S4, the detection result is stored and conveyed, and corona discharge signal collection comprises three parts of high-frequency signal monitoring, ultraviolet monitoring and characteristic product monitoring. According to the method, three different corona discharge signals of high-frequency signals, ultraviolet rays and characteristic products are integrated, and the corona discharge of the end part of the stator winding is comprehensively judged, so that the problems of limitation, easiness in interference and the like existing in a traditional single-factor detection method are solved; and an accurate judgment basis is provided for corona discharge detection of the end part of the stator winding. A high-frequency sensor of an improved composite antenna is selected to monitor high-frequency signals, and the width of a monitoring frequency band is expanded.

Description

Technical Field

[0001] The invention relates to the technical field of corona discharge detection of stator winding ends of large generators, and more specifically to a corona discharge detection method of stator winding ends based on multi-signal joint monitoring. Background Art

[0002] Corona discharge is a self-sustaining discharge phenomenon unique to extremely non-uniform electric fields. Electrodes with a small radius of curvature will produce corona discharge under the action of a non-uniform field that exceeds the corona initiation voltage. In actual operation, the electric field distribution on the surface of the stator winding of a large generator is uneven, so corona discharge is prone to occur. The stator winding end has a smaller radius of curvature and a smaller wire spacing, so the electric field non-uniformity is more obvious. Under the action of the non-uniform electric field, the corona discharge problem is often the most serious. Corona discharge will produce characteristic products such as ozone (O3), nitrogen oxides (NO, NO2), hydrogen (H2), hydrocarbon compounds, and ketone compounds, which will corrode the insulation at the stator end and promote insulation aging. Long-term corona discharge will lead to insulation aging, and in severe cases, even cause accidents, endangering the stable operation of the power system. In order to maintain the normal operation of the generator, the stator winding should be treated with anti-corona treatment in time. Most of the treatment methods include: cleaning the dirt and carbides on the winding surface, re-applying anti-corona paint, and rewinding or wrapping the insulation according to the actual situation.

[0003] Timely treatment of corona discharge at the end of stator windings of large generator sets is the key to improving the stability of system operation. Therefore, it is of great significance to timely and accurately detect and evaluate the corona discharge of windings and transmit information to the management side so that necessary anti-corona measures can be taken in time. According to the characteristics of corona discharge, visible light waves, ultraviolet rays, sound waves, high-frequency electromagnetic waves will be radiated, and characteristic products such as ozone and nitrogen oxides will be produced. At present, common corona detection methods include visual inspection method and ultraviolet detection method. In addition, there are corona current detection method, high-frequency signal detection method, electromagnetic wave detection method, sound wave detection method, ozone detection method, etc. However, the existing detection methods have the disadvantages of low gain, large transmission loss, low efficiency, etc., and cannot completely eliminate the interference of ultraviolet light in the environment. There are also problems of ultraviolet light absorption and scattering, and there are also limitations in corona detection.

[0004] Traditional corona detection methods often have certain limitations and instabilities, and are easily interfered by environmental factors, and the detection results are accidental. Therefore, how to overcome the limitations of traditional single-factor corona detection, provide a comprehensive and integrated criterion for stator winding end corona detection, and improve the accuracy and sensitivity of the judgment results has become an urgent problem that researchers in this field need to solve. Summary of the invention

[0005] In order to overcome the defects in the above-mentioned prior art, the present invention discloses a stator winding end corona discharge detection method based on multi-signal joint monitoring to solve the limitations and susceptibility to interference of the traditional single-factor corona detection method. The present invention combines three different corona discharge signals, namely high-frequency signals, ultraviolet rays and characteristic products, to make a comprehensive and integrated judgment on the corona discharge at the end of the stator winding, thus overcoming the limitations and susceptibility to interference of the traditional single-factor detection method, and providing a comprehensive and accurate diagnostic method for the detection of corona discharge at the end of the stator winding.

[0006] In order to achieve the above objectives, the technical solution adopted by the present invention is:

[0007] A method for detecting corona discharge at the end of a stator winding based on multi-signal joint monitoring comprises the following steps:

[0008] 1. Corona discharge signal acquisition

[0009] Step S1, collecting high-frequency signals, ultraviolet signals and characteristic product contents at the ends of the stator windings, as well as environmental influencing factors of the stator windings;

[0010] Preferably, in step S1, the environmental influencing factors include detection distance, temperature, humidity and air pressure of the stator winding working environment.

[0011] Step S1 collects corona discharge signals: comprehensively collects possible signals of corona discharge at the end of the stator winding, monitors and collects high-frequency signals, ultraviolet signals and characteristic product contents at the end of the stator winding respectively, and collects environmental influencing factors of the stator winding, including detection distance, temperature, humidity and air pressure of the working environment of the stator winding.

[0012] Preferably, step S1 comprises:

[0013] The high-frequency signal adopts a high-frequency wireless detection method and uses a high-frequency sensor for monitoring, wherein the radiation surface of the high-frequency sensor antenna is a composite antenna of an Archimedean spiral curve and an equiangular spiral curve;

[0014] The ultraviolet signal adopts the solar blind zone ultraviolet imaging detection method, and is monitored using an ultraviolet imager with the imaging gain set to 40%-80%;

[0015] Ozone and nitrogen oxides are selected as characteristic product contents for detection, and the detection instruments adopt ozone semiconductor sensors and nitrogen oxide semiconductor sensors.

[0016] In step S1, the high-frequency signal adopts a high-frequency wireless detection method and is monitored by a high-frequency sensor, wherein the radiation surface of the high-frequency sensor antenna is a composite antenna of an Archimedean spiral curve and an equiangular spiral curve to increase the frequency bandwidth of the sensor. The ultraviolet signal adopts a solar blind zone ultraviolet imaging detection method and is monitored by an ultraviolet imager. The imaging gain is preferably set to 40%-80% to make the image radiate and improve the sensitivity of the ultraviolet imager. Ozone and nitrogen oxides are selected as characteristic products for detection, and the detection instruments use ozone semiconductor sensors and nitrogen oxide semiconductor sensors.

[0017] Preferably, in step S1, the radiation surface of the high-frequency sensor antenna is a composite antenna of an Archimedean spiral curve and an equiangular spiral curve:

[0018] The outer curve is an Archimedean spiral curve, and the expression is

[0019] The composite antenna whose inner curve is an equiangular spiral curve can be expressed as

[0020] Where r is the distance from any point on the spiral to the origin; is the azimuth of the point; is the starting angle of the spiral; r0 is the starting radius of the spiral; k is the spiral growth rate, which is a constant.

[0021] 2. Corona discharge signal processing

[0022] Step S2, setting a corona discharge threshold and a characteristic frequency band according to the collected environmental impact factors, and comparing the collected high-frequency signal, ultraviolet signal and characteristic product content with the set corona discharge threshold and characteristic frequency band, respectively, to determine the corona detection result based on the high-frequency signal, ultraviolet signal and characteristic product content;

[0023] Step S2 processes the corona discharge signal: according to the collected environmental impact factors and the actual measurement situation, the corona discharge threshold and characteristic frequency band are set, the collected three types of signals are compared with the set corona discharge threshold and characteristic frequency band, and the corona detection result based on each type of signal is determined.

[0024] In the present invention, in the corona discharge threshold and the characteristic frequency band, the corona discharge threshold includes a corona discharge fluctuation threshold, a corona discharge ultraviolet photon number threshold, a corona discharge ozone volume fraction threshold and a corona discharge nitrogen oxide volume fraction threshold.

[0025] Preferably, step S2 comprises:

[0026] When the high-frequency signal produces obvious fluctuations in the set corona characteristic frequency band and exceeds the set corona discharge fluctuation threshold, a high-frequency over-limit signal is output;

[0027] The number of ultraviolet photons in the ultraviolet signal is compared with the set corona discharge ultraviolet photon number threshold, and the number of ultraviolet photons monitored at different motor stator ends under the same operating conditions is compared. When the number of ultraviolet photons exceeds the corona discharge ultraviolet photon number threshold, or exceeds the allowable deviation value with the average monitoring value of other motor stators, an ultraviolet over-limit signal is output;

[0028] When the ozone volume fraction in the characteristic product content exceeds the set corona discharge ozone volume fraction threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge nitrogen oxide volume fraction threshold, a characteristic product over-limit signal is output.

[0029] In the above step S2, when the high-frequency signal in the storage generates obvious fluctuations in the set corona characteristic frequency band and exceeds the set fluctuation threshold, a high-frequency over-limit signal is output. The number of ultraviolet photons monitored in the storage is compared with the set corona discharge threshold, and the number of ultraviolet photons monitored at different motor stator ends under the same operating conditions is compared. When the number of ultraviolet photons exceeds the corona discharge threshold, or the average monitoring value with other motor stators exceeds the allowable deviation value, an ultraviolet over-limit signal is output. When the ozone volume fraction in the storage exceeds the set corona discharge threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge threshold, a characteristic product over-limit signal is output.

[0030] 3. Corona discharge status judgment

[0031] Step S3, using the high-frequency signal, ultraviolet signal and corona detection results of characteristic product content to comprehensively judge the corona discharge state and output it;

[0032] Step S3 is to judge the state of corona discharge: a comprehensive judgment of the state of corona discharge is made according to the corona detection results of the three types of signals.

[0033] Preferably, in step S3, the comprehensive determination of the corona discharge state includes:

[0034] When two or more types of judgment results are output as over-limit signals, corona discharge occurs at the stator winding end and a corona discharge signal is output;

[0035] When a type of judgment result is an output over-limit signal, the output re-collects and detects the high-frequency signal, ultraviolet signal and characteristic product content feature items;

[0036] When no over-limit signal is output, no corona discharge occurs at the stator winding end and no corona discharge signal is output.

[0037] The comprehensive judgment method of the corona discharge state in the above step S3 is: when two or more types of judgment results output an over-limit signal, the comprehensive judgment of the state outputs a corona discharge signal; when one type of judgment result outputs an over-limit signal, the characteristic items are re-measured; when no over-limit signal is output, no corona discharge signal is output.

[0038] 4. Storage and delivery of test results

[0039] Step S4: store the corona detection results and the comprehensive judgment results, and transmit them to the management personnel end.

[0040] Step S4 stores and transmits the detection results: the signal collection results and status judgment results are stored and transmitted to the management personnel end.

[0041] Preferably, in step S4, the management end displays the comprehensive judgment result of corona discharge at the stator winding end, as well as the corona detection results of the high-frequency signal fluctuation frequency band, ultraviolet photon number, ozone volume fraction and nitrogen oxide volume fraction at the stator winding end.

[0042] In step S4, the comprehensive judgment result of corona discharge at the end of the stator winding, as well as the detection results of the high-frequency signal fluctuation frequency band, ultraviolet photon number, ozone and nitrogen oxide volume fraction at the end of the stator winding can be displayed on the management side to facilitate the management to perform corresponding management.

[0043] The operation steps of the present invention are as follows:

[0044] (1) Initialize the data of the detection system and set the sampling rate and gain after initialization.

[0045] (2) Collect environmental parameters, including detection distance, temperature, humidity, and air pressure of the stator winding working environment.

[0046] (3) Comprehensively set the discharge threshold and characteristic frequency band based on the collected environmental parameters;

[0047] (4) Conduct corona discharge multi-signal detection, including high-frequency signal detection, ultraviolet signal detection, ozone and nitrogen oxide content detection, and input the detection results into a storage device.

[0048] (5) Compare the detection results of the three types of discharge signals in the storage device with the set discharge threshold or characteristic frequency band. When the high-frequency signal produces obvious fluctuations in the set corona characteristic frequency band and exceeds the set fluctuation threshold, a high-frequency over-limit signal is output; the number of ultraviolet photons is compared with the set corona discharge threshold, and the number of ultraviolet photons monitored at the ends of different motor stators under the same operating conditions are compared with each other. When the number of ultraviolet photons exceeds the corona discharge threshold, or exceeds the allowable deviation value with the average monitoring value of other motor stators, an ultraviolet over-limit signal is output; when the ozone volume fraction exceeds the set corona discharge threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge threshold, a characteristic product over-limit signal is output.

[0049] (6) Perform a comprehensive judgment on corona discharge. If no possible discharge signal is output in step (5), it is considered that no corona discharge has occurred and the detection process ends; when one type of judgment result outputs an over-limit signal, the characteristic item is re-measured; when two or more types of judgment results output an over-limit signal, the state comprehensive judgment outputs a corona discharge signal and transmits the comprehensive detection data to the management end.

[0050] Preferably, in step S1 and step S2, the high-frequency signal acquisition detection includes the following steps:

[0051] A1. Start the high-frequency sensor and set the sampling rate;

[0052] A2. High-frequency signal monitoring: The high-frequency sensor antenna receives the high-frequency electromagnetic wave signal from the end of the stator winding, the impedance converter performs impedance matching on the received high-frequency electromagnetic wave signal, the microwave absorption back cavity of the high-frequency sensor unidirectionally radiates the processed electromagnetic wave signal, and then the dielectric coating cover plate and the low-noise RF amplifier perform gain amplification and secondary signal amplification on the radiated signal;

[0053] A3, the low-noise RF amplifier inputs the amplified signal into the high-speed digital oscilloscope and storage to collect and store the corona discharge high-frequency signal;

[0054] A4. Compare the high-frequency signal monitoring value in the storage with the set characteristic frequency band. When the high-frequency signal in the storage produces obvious fluctuations in the set corona characteristic frequency band and exceeds the set fluctuation threshold, a high-frequency over-limit signal is output and the high-frequency signal detection process ends; otherwise, no signal is output and the high-frequency signal detection process ends.

[0055] Preferably, in step S1 and step S2, the ultraviolet signal collection and detection includes the following steps:

[0056] B1. Start the UV imager, set the sampling rate and filter switching speed, and set the imaging gain;

[0057] B2. Ultraviolet signal monitoring: Switch the ultraviolet imager to the solar blind filter to monitor the ultraviolet light signal that may be radiated by corona discharge at the end of the stator winding; switch the ultraviolet imager to the ultraviolet cutting filter to obtain an image of the working condition of the stator winding end under visible light; overlap the images obtained in the two cases to count ultraviolet photons and perform data analysis and calibration;

[0058] B3, inputting the ultraviolet signal monitoring result into a storage device to store the corona discharge ultraviolet signal;

[0059] B4. Compare the ultraviolet signal monitoring value in the storage with the set characteristic frequency band. When the number of ultraviolet photons exceeds the corona discharge ultraviolet photon number threshold, or the average monitoring value of ultraviolet photons of other motor stators exceeds the allowable deviation value, an ultraviolet over-limit signal is output and the ultraviolet signal detection process ends; otherwise, no signal is output and the ultraviolet signal detection process ends.

[0060] Preferably, in step S1 and step S2, the characteristic product content collection and detection includes the following steps:

[0061] C1. Start the ozone semiconductor sensor and the nitrogen oxide semiconductor sensor and set the parameters;

[0062] C2. Monitor the ozone and nitrogen oxide contents by using an ozone semiconductor sensor and a nitrogen oxide semiconductor sensor, and input the detected real-time contents of ozone and nitrogen oxide at the stator winding end into a storage;

[0063] C3. Compare the ozone and nitrogen oxide contents in the storage with the set content thresholds of the two gases respectively. When the ozone volume fraction exceeds the set corona discharge ozone volume fraction threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge nitrogen oxide volume fraction threshold, a characteristic product over-limit signal is output and the characteristic product detection process ends; otherwise, no signal is output and the characteristic product detection process ends.

[0064] Preferably, in step S1, step S2, step S3 and step S4, the comprehensive detection process includes the following steps:

[0065] D1. Start the detection equipment, initialize the equipment, and set the acquisition parameters of the three characteristic signals: high-frequency signal, ultraviolet signal, and characteristic product content;

[0066] D2. Collect environmental influencing factors, including detection distance, temperature, humidity and air pressure of the stator winding working environment;

[0067] D3. Comprehensively set the high-frequency signal characteristic frequency band and allowable fluctuation threshold, ultraviolet photon number threshold, ozone volume fraction threshold, and nitrogen oxide volume fraction threshold of the stator winding end corona discharge based on the collected environmental impact factors;

[0068] D4. Conduct high frequency signal, ultraviolet signal, ozone and nitrogen oxide detection, and store the detection results in the storage;

[0069] D5. Compare the detection results of the three characteristic signals with the set thresholds and characteristic frequency bands respectively, and determine whether to output an over-limit signal according to the comparison results;

[0070] D6, detect whether the output is an over-limit signal:

[0071] If no over-limit signal is detected, no corona discharge signal is output, and the test result is transmitted to the management end and displayed, and the comprehensive test process ends;

[0072] If two or more characteristic signal detection outputs are detected, an over-limit signal is output, a corona discharge signal is output, and the detection results are transmitted to the management end and displayed, and the comprehensive detection process ends;

[0073] If it is detected that there is a characteristic product detection output over-limit signal, the three characteristic signals are detected again.

[0074] Beneficial effects of the present invention:

[0075] The stator winding end corona discharge detection method provided by the present invention comprises: step S1, collecting corona discharge signals; step S2, processing corona discharge signals; step S3, judging the state of corona discharge; step S4, storing and transmitting the detection results, wherein the collection of corona discharge signals comprises three parts: high-frequency signal monitoring, ultraviolet monitoring and characteristic product monitoring. The present invention combines three different corona discharge signals, namely high-frequency signals, ultraviolet rays and characteristic products, to make a comprehensive and integrated judgment on the corona discharge at the stator winding end, overcomes the limitations and susceptibility to interference of the traditional single-factor detection method, and provides an accurate judgment basis for the detection of corona discharge at the stator winding end.

[0076] The present invention proposes a stator winding end corona discharge detection method based on multi-signal joint monitoring, which overcomes the limitations, instability, and susceptibility to interference from environmental factors of the existing stator winding end corona detection method based on a single characteristic factor, resulting in randomness in the detection results, and integrates three different corona discharge signals of high-frequency signals, ultraviolet rays, and characteristic products to make a comprehensive and integrated judgment on the stator winding end corona discharge. The present invention selects a high-frequency sensor with an improved composite antenna to monitor high-frequency signals, expanding the width of the monitoring frequency band; uses an ultraviolet imager to detect ultraviolet signals, and realizes quantitative analysis by counting ultraviolet photons and data analysis calibration; and selects ozone and nitrogen oxides as characteristic products at the same time, avoiding inaccurate judgment results caused by nitrogen oxides produced by corona promoting ozone decomposition, and comprehensively improving the accuracy and sensitivity of corona detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a schematic diagram of the detection method of the present invention;

[0078] Figure 2 It is a schematic diagram of the high-frequency signal detection process of the present invention;

[0079] Figure 3 It is a schematic diagram of the ultraviolet signal detection process of the present invention;

[0080] Figure 4 It is a schematic diagram of the characteristic product detection process of the present invention;

[0081] Figure 5 It is a schematic diagram of the operation flow of the present invention. DETAILED DESCRIPTION

[0082] The concept, specific structure and technical effects of the present invention will be clearly and completely described below in conjunction with the embodiments and drawings to fully understand the purpose, characteristics and effects of the present invention.

[0083] Example 1

[0084] A method for detecting corona discharge at the end of stator winding based on multi-signal joint monitoring, such as Figure 1 As shown, the following steps are included:

[0085] Step S1, signal acquisition: Comprehensively collect possible signals of corona discharge at the end of the stator winding, and monitor and collect high-frequency signals, ultraviolet signals and volume fractions of characteristic products at the end of the stator winding. Specifically, the high-frequency signal adopts a high-frequency wireless detection method and a high-frequency sensor for monitoring; the ultraviolet signal adopts a solar blind zone ultraviolet imaging detection method and an ultraviolet imager for monitoring; ozone and nitrogen oxides are selected as characteristic products for detection, and the detection instruments adopt ozone semiconductor sensors and nitrogen oxide semiconductor sensors.

[0086] Step S2, signal processing: compare the three types of signals collected with the corona discharge threshold and characteristic frequency band set based on the actual standard, and judge the corona detection results based on each type of signal. Specifically, when the high-frequency signal in the storage device produces obvious fluctuations in the set corona characteristic frequency band and exceeds the set fluctuation threshold, a high-frequency over-limit signal is output; the number of ultraviolet photons monitored in the storage device is compared with the set corona discharge threshold, and the number of ultraviolet photons monitored at the ends of different motor stators under the same operating conditions is compared. When the number of ultraviolet photons exceeds the corona discharge threshold, or the average monitoring value with other motor stators exceeds the allowable deviation value, an ultraviolet over-limit signal is output; when the ozone volume fraction in the storage device exceeds the set corona discharge threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge threshold, a characteristic product over-limit signal is output. A single-chip microcomputer system can be used for signal processing, and the embodiment of the present invention does not limit this.

[0087] Step S3, state judgment: perform a comprehensive judgment on the corona discharge state according to the corona detection results of the three types of signals. Specifically, when two or more types of judgment results output an over-limit signal, the state comprehensive judgment outputs a corona discharge signal; when one type of judgment result outputs an over-limit signal, the characteristic item is re-measured; when no over-limit signal is output, no corona discharge signal is output. The state judgment can use a single-chip microcomputer system, which is not limited in the embodiment of the present invention.

[0088] Step S4, storage and transmission: the signal acquisition results and the state judgment results are stored, and the comprehensive judgment results of the corona discharge at the stator winding end, the high-frequency signal fluctuation frequency band at the stator winding end, the number of ultraviolet photons, and the ozone and nitrogen oxide volume fraction detection results are transmitted to the management end for display. The storage and transmission method can use Internet mobile communication, which is not limited in the embodiment of the present invention.

[0089] Example 2

[0090] This embodiment is further described on the basis of embodiment 1. Figure 2 As shown, the high-frequency signal detection process includes the following steps:

[0091] (1) Start the high-frequency sensor and set the sampling rate.

[0092] (2) High-frequency signal monitoring: The sensor antenna receives the high-frequency electromagnetic wave signal from the end of the stator winding; the impedance transformer performs impedance matching on the received high-frequency electromagnetic wave signal; the microwave absorption back cavity of the sensor unidirectionally radiates the processed electromagnetic wave signal; the dielectric coating cover plate and the low-noise RF amplifier perform gain amplification and signal secondary amplification on the radiated signal;

[0093] (3) The low-noise RF amplifier inputs the amplified signal into a high-speed digital oscilloscope and storage to realize the collection and storage of the corona discharge high-frequency signal.

[0094] (4) Compare the high-frequency signal monitoring value in the storage device with the set characteristic frequency band. When the high-frequency signal in the storage device produces obvious fluctuations in the set corona characteristic frequency band and exceeds the set fluctuation threshold, a high-frequency over-limit signal is output and the high-frequency signal detection process ends. Otherwise, no signal is output and the high-frequency signal detection process ends.

[0095] The high-frequency sensor antenna is a composite antenna whose radiation surface is an Archimedean spiral curve and an equiangular spiral curve; the outer curve is an Archimedean spiral curve, and the expression is

[0096] The composite antenna whose inner curve is an equiangular spiral curve can be expressed as

[0097] Where r is the distance from any point on the spiral to the origin; is the azimuth of the point; is the starting angle of the spiral; r0 is the starting radius of the spiral; k is the spiral growth rate, which is a constant.

[0098] Example 3

[0099] This embodiment is further described on the basis of embodiment 1. Figure 3 As shown, the UV signal detection process includes the following steps:

[0100] (1) Start the UV imager, set the sampling rate and filter switching speed, and set the appropriate imaging gain.

[0101] (2) Ultraviolet signal monitoring: Switch the ultraviolet light detector to the solar blind zone filter to monitor the ultraviolet light signal that may be radiated by corona discharge at the end of the stator winding; switch to the ultraviolet light cutting filter to obtain an image of the working condition of the stator winding end under visible light; overlap the images obtained in the two cases to count ultraviolet photons and calibrate the data analysis.

[0102] (3) Inputting the ultraviolet signal monitoring result into the storage device to realize the storage of the corona discharge ultraviolet signal.

[0103] (4) Compare the ultraviolet signal monitoring value in the storage device with the set characteristic frequency band. When the number of ultraviolet photons exceeds the corona discharge threshold, or when the average monitoring value of ultraviolet photons of other motor stators exceeds the allowable deviation value, an ultraviolet over-limit signal is output and the ultraviolet signal detection process ends. Otherwise, no signal is output and the ultraviolet signal detection process ends.

[0104] Example 4

[0105] This embodiment is further described on the basis of embodiment 1. Figure 4 As shown, the characteristic product detection process includes the following steps:

[0106] (1) Start the ozone semiconductor sensor and the nitrogen oxide semiconductor sensor and set the parameters;

[0107] (2) Monitor the ozone and nitrogen oxide content, and input the real-time content of ozone and nitrogen oxide at the stator winding end obtained by detection into a storage device.

[0108] (3) Compare the ozone and nitrogen oxide contents in the storage device with the set content thresholds of the two gases respectively. When the ozone volume fraction exceeds the set corona discharge threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge threshold, a characteristic gas over-limit signal is output, and the characteristic product detection process ends; otherwise, no signal is output, and the characteristic product detection process ends.

[0109] Example 5

[0110] This embodiment is further described on the basis of embodiment 1. Figure 5 As shown, the comprehensive detection process includes the following steps:

[0111] (1) Start the detection equipment, initialize the equipment, and set the acquisition parameters of the three characteristic signals respectively.

[0112] (2) Collect environmental influencing factors, including detection distance, temperature, humidity and air pressure of the stator winding working environment.

[0113] (3) Based on the collected environmental impact factors, the high-frequency signal characteristic frequency band and the allowable fluctuation threshold, ultraviolet photon number threshold, ozone volume fraction threshold, and nitrogen oxide volume fraction threshold of the stator winding end corona discharge are comprehensively set.

[0114] (4) Conduct high-frequency signal, ultraviolet signal, ozone and nitrogen oxide detection, and store the detection results in a storage device.

[0115] (5) The detection results of the three characteristic signals are compared with the set thresholds and characteristic frequency bands respectively, and whether to output an out-of-limit signal is determined according to the comparison results.

[0116] (6) Detect whether an over-limit signal is output: If no over-limit signal is detected, no corona discharge signal is output, and the detection result is transmitted to the management end and displayed, and the comprehensive detection process ends; if it is detected that two or more characteristic signal detections have output over-limit signals, a corona discharge signal is output, and the detection result is transmitted to the management end and displayed, and the comprehensive detection process ends; if it is detected that one characteristic product detection has output an over-limit signal, the three characteristic signals are re-detected.

[0117] The above is a specific description of the implementation mode of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention, and these equivalents or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A stator winding end corona discharge detection method based on multi-signal joint monitoring, characterized in that: The following steps are involved: Step S1, collecting high-frequency signals, ultraviolet signals and characteristic product contents at the ends of the stator windings, as well as environmental influencing factors of the stator windings; Step S2, setting a corona discharge threshold and a characteristic frequency band according to the collected environmental impact factors, and comparing the collected high-frequency signal, ultraviolet signal and characteristic product content with the set corona discharge threshold and characteristic frequency band, respectively, to determine the corona detection result based on the high-frequency signal, ultraviolet signal and characteristic product content; Step S3, using the high-frequency signal, ultraviolet signal and corona detection results of characteristic product content to comprehensively judge the corona discharge state and output it; Step S4: store the corona detection results and the comprehensive judgment results, and transmit them to the management personnel end.

2. The stator winding end corona discharge detection method according to claim 1, characterized in that: Step S1 includes: The high-frequency signal adopts a high-frequency wireless detection method and uses a high-frequency sensor for monitoring, wherein the radiation surface of the high-frequency sensor antenna is a composite antenna of an Archimedean spiral curve and an equiangular spiral curve; The ultraviolet signal adopts the solar blind zone ultraviolet imaging detection method, and is monitored using an ultraviolet imager with the imaging gain set to 40%-80%; Ozone and nitrogen oxides are selected as characteristic product contents for detection, and the detection instruments adopt ozone semiconductor sensors and nitrogen oxide semiconductor sensors.

3. The stator winding end corona discharge detection method according to claim 2, characterized in that: In step S1, the radiation surface of the high-frequency sensor antenna is a composite antenna of an Archimedean spiral curve and an equiangular spiral curve: The outer curve is an Archimedean spiral curve, and the expression is The composite antenna whose inner curve is an equiangular spiral curve can be expressed as Where r is the distance from any point on the spiral to the origin; is the azimuth of the point; is the spiral starting angle; r0 is the spiral starting radius; k is the spiral growth rate.

4. The stator winding end corona discharge detection method according to claim 1, characterized in that: Step S2 includes: When the high-frequency signal produces obvious fluctuations in the set corona characteristic frequency band and exceeds the set corona discharge fluctuation threshold, a high-frequency over-limit signal is output; The number of ultraviolet photons in the ultraviolet signal is compared with the set corona discharge ultraviolet photon number threshold, and the number of ultraviolet photons monitored at different motor stator ends under the same operating conditions is compared. When the number of ultraviolet photons exceeds the corona discharge ultraviolet photon number threshold, or exceeds the allowable deviation value with the average monitoring value of other motor stators, an ultraviolet over-limit signal is output; When the ozone volume fraction in the characteristic product content exceeds the set corona discharge ozone volume fraction threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge nitrogen oxide volume fraction threshold, a characteristic product over-limit signal is output.

5. The stator winding end corona discharge detection method according to claim 1, characterized in that: In step S3, the comprehensive determination of the corona discharge state includes: When two or more types of judgment results are output as over-limit signals, corona discharge occurs at the stator winding end and a corona discharge signal is output; When a type of judgment result is an output over-limit signal, the output re-collects and detects the high-frequency signal, ultraviolet signal and characteristic product content feature items; When no over-limit signal is output, no corona discharge occurs at the stator winding end and no corona discharge signal is output.

6. The stator winding end corona discharge detection method according to claim 1, characterized in that: In step S4, the management terminal displays the comprehensive judgment result of the corona discharge at the stator winding end, as well as the corona detection results of the high-frequency signal fluctuation frequency band, ultraviolet photon number, ozone volume fraction and nitrogen oxide volume fraction at the stator winding end.

7. The stator winding end corona discharge detection method according to claim 1, characterized in that: In step S1 and step S2, the high-frequency signal acquisition detection includes the following steps: A1. Start the high-frequency sensor and set the sampling rate; A2. High-frequency signal monitoring: The high-frequency sensor antenna receives the high-frequency electromagnetic wave signal from the end of the stator winding, the impedance converter performs impedance matching on the received high-frequency electromagnetic wave signal, the microwave absorption back cavity of the high-frequency sensor unidirectionally radiates the processed electromagnetic wave signal, and then the dielectric coating cover plate and the low-noise RF amplifier perform gain amplification and secondary signal amplification on the radiated signal; A3, the low-noise RF amplifier inputs the amplified signal into the high-speed digital oscilloscope and storage to collect and store the corona discharge high-frequency signal; A4. Compare the high-frequency signal monitoring value in the storage with the set characteristic frequency band. When the high-frequency signal in the storage produces obvious fluctuations in the set corona characteristic frequency band and exceeds the set fluctuation threshold, a high-frequency over-limit signal is output and the high-frequency signal detection process ends; otherwise, no signal is output and the high-frequency signal detection process ends.

8. The stator winding end corona discharge detection method according to claim 1, characterized in that: In step S1 and step S2, the ultraviolet signal collection and detection includes the following steps: B1. Start the UV imager, set the sampling rate and filter switching speed, and set the imaging gain; B2. Ultraviolet signal monitoring: Switch the ultraviolet imager to the solar blind filter to monitor the ultraviolet light signal that may be radiated by corona discharge at the end of the stator winding; switch the ultraviolet imager to the ultraviolet cutting filter to obtain an image of the working condition of the stator winding end under visible light; overlap the images obtained in the two cases to count ultraviolet photons and perform data analysis and calibration; B3, inputting the ultraviolet signal monitoring result into a storage device to store the corona discharge ultraviolet signal; B4. Compare the ultraviolet signal monitoring value in the storage with the set characteristic frequency band. When the number of ultraviolet photons exceeds the corona discharge ultraviolet photon number threshold, or the average monitoring value of ultraviolet photons of other motor stators exceeds the allowable deviation value, an ultraviolet over-limit signal is output and the ultraviolet signal detection process ends; otherwise, no signal is output and the ultraviolet signal detection process ends.

9. The stator winding end corona discharge detection method according to claim 1, characterized in that: In step S1 and step S2, the characteristic product content collection and detection includes the following steps: C1. Start the ozone semiconductor sensor and the nitrogen oxide semiconductor sensor and set the parameters; C2. Monitor the ozone and nitrogen oxide contents by using an ozone semiconductor sensor and a nitrogen oxide semiconductor sensor, and input the detected real-time contents of ozone and nitrogen oxide at the stator winding end into a storage; C3. Compare the ozone and nitrogen oxide contents in the storage with the set content thresholds of the two gases respectively. When the ozone volume fraction exceeds the set corona discharge ozone volume fraction threshold, or the nitrogen oxide volume fraction exceeds the set corona discharge nitrogen oxide volume fraction threshold, a characteristic product over-limit signal is output and the characteristic product detection process ends; otherwise, no signal is output and the characteristic product detection process ends.

10. The stator winding end corona discharge detection method according to claim 1, characterized in that: In step S1, step S2, step S3 and step S4, the comprehensive detection process includes the following steps: D1. Start the detection equipment, initialize the equipment, and set the acquisition parameters of the three characteristic signals: high-frequency signal, ultraviolet signal, and characteristic product content; D2. Collect environmental influencing factors, including detection distance, temperature, humidity and air pressure of the stator winding working environment; D3. Comprehensively set the high-frequency signal characteristic frequency band and allowable fluctuation threshold, ultraviolet photon number threshold, ozone volume fraction threshold, and nitrogen oxide volume fraction threshold of the stator winding end corona discharge based on the collected environmental impact factors; D4. Conduct high frequency signal, ultraviolet signal, ozone and nitrogen oxide detection, and store the detection results in the storage; D5. Compare the detection results of the three characteristic signals with the set thresholds and characteristic frequency bands respectively, and determine whether to output an over-limit signal according to the comparison results; D6, detect whether the output is an over-limit signal: If no over-limit signal is detected, no corona discharge signal is output, and the test result is transmitted to the management end and displayed, and the comprehensive test process ends; If two or more characteristic signal detection outputs are detected, an over-limit signal is output, a corona discharge signal is output, and the detection results are transmitted to the management end and displayed, and the comprehensive detection process ends; If it is detected that there is a characteristic product detection output over-limit signal, the three characteristic signals are detected again.

Images