Multi-source partial discharge detection device and positioning method in high-voltage switchgear
By using a combination device of a broadband sensor and a host in a high-voltage switch cabinet, using time window and mutual correlation coefficient to process signals, the precise positioning of multi-source local distribution signals is achieved, and the positioning of multi-source local distribution signals in the existing technology is solved. Accurate question.
Patent Information
- Application Number
- CN202211289411.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-10-20
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-10-20
AI Technical Summary
The prior art is difficult to accurately locate multi-source local discharge signals in high-voltage switch cabinets, especially when signals are mixed, resulting in inaccurate positioning.
Using a detection device composed of a broadband sensor and a host, signals are collected and processed through a microprocessor, filtering and amplification unit and high-frequency data acquisition unit, and signal pairing and positioning are achieved by using time windows and mutual correlation coefficients to realize multi-source local signal release accurate positioning.
Accurate positioning of multiple local and distributed sources in high-voltage switch cabinets is achieved, detection accuracy is improved, and positioning difficulties under multi-source local and distributed signals is solved.
Smart Images

Figure CN115616357B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electric power detection, and in particular to a multi-source partial discharge detection device and a positioning method in a high-voltage switch cabinet. Background Art
[0002] High-voltage switches refer to electrical equipment that plays the role of switching, controlling or protecting in power generation, transmission, distribution, power conversion and consumption of power systems. They mainly include high-voltage circuit breakers, high-voltage disconnectors and earthing switches, high-voltage load switches, high-voltage automatic reclosing and sectionalizers, high-voltage operating mechanisms, high-voltage explosion-proof distribution devices and high-voltage switchgear, etc. They play a vital role in the safe operation of power grids and the distribution and metering of electric energy.
[0003] In view of this, the monitoring and diagnosis of the operating status of high-voltage switchgear has been a topic of concern for scientific research institutes, monitoring equipment manufacturers, power production and power grid companies for many years, and a variety of monitoring technologies have been proposed. For example, by collecting partial discharge signals at multiple points and then using the traveling wave method to locate switchgear with partial discharge problems, however, the existing technology simply uses propagation time and propagation speed to locate the position of a single-point partial discharge, ignoring the fact that the signal collected at the collection point may be a mixed signal of multi-source partial discharge, and therefore, it is impossible to accurately locate the source of partial discharge. Summary of the invention
[0004] In view of this, the present invention provides a multi-source partial discharge detection device and positioning method in a high-voltage switch cabinet, which can accurately locate multiple partial discharge sources based on multi-source partial discharge mixed signals.
[0005] The technical solution adopted by the embodiment of the present invention to solve the technical problem is:
[0006] A multi-source partial discharge detection device in a high-voltage switch cabinet, characterized in that it comprises: two broadband sensors and a host, wherein the two broadband sensors serve as a master-end sensor and a slave-end sensor respectively, and the host comprises a microprocessor, a high-frequency data acquisition unit, a filtering and amplifying unit, and a power supply unit;
[0007] The broadband sensor comprises a needle electrode, an insulating layer plate, a circular plate electrode and a circuit board which are arranged in sequence from top to bottom: a through hole matching the size of the needle electrode is provided in the center of the insulating layer plate and is located on the lower surface of the insulating layer plate, and the edge of the through hole has an annular protrusion; the needle electrode is inserted into the through hole, the needle electrode is fixedly connected to the inner surface of the through hole, the head of the needle electrode is used to sense an electric field signal, and the end of the needle electrode is connected to the positive input terminal of the circuit board; a through hole is provided in the center of the plate electrode, the upper surface of the plate electrode is fixedly connected to the lower surface of the insulating layer plate, and the annular protrusion of the insulating layer plate passes through the through hole; the plate electrode is used to sense an electric field signal, The plate electrode and the circuit board are fixedly connected via a hollow plate electrode connector, and the plate electrode connector is electrically connected to the negative input terminal and the ground terminal of the circuit board; the needle electrode and the plate electrode connector are coaxially arranged, and the distance between the needle electrode and the inner wall of the plate electrode connector is d. The needle electrode serves as an inner conductor, the plate electrode and the plate electrode connector serve as an outer conductor, and the insulating layer plate serves as an insulating medium, and the three are combined to form a coaxial feeding form; the circuit board is powered by a battery, and the circuit board has the positive input terminal, the negative input terminal, the ground terminal and the output terminal, and the output terminal of the circuit board is connected to the host through a coaxial transmission line, and the output signal U of the circuit board out is the needle electrode sensing signal U after conditioning and amplification N The plate electrode sensing signal U P The result obtained by the signal difference between
[0008] The high-frequency data acquisition unit is used to acquire output signals of the two broadband sensors through an acquisition card;
[0009] The filtering and amplifying unit performs filtering and noise reduction on the output signals of the two broadband sensors;
[0010] The microprocessor is connected to the high-frequency data acquisition unit and the filter amplification unit, and is used to trigger the high-frequency data acquisition unit to collect signals; receive the signals processed by the filter amplification unit and send them to the computer terminal;
[0011] The power supply unit is used to supply power to the microprocessor, the high-frequency data acquisition unit and the filtering and amplifying unit.
[0012] Preferably, the broadband sensor collects partial discharge signals in a frequency band ranging from 1 MHz to 200 MHz, and the output signal U of the broadband sensor is out And the partial discharge signal U in the switch cabinet sig The corresponding relationship is:
[0013]
[0014] The above formula is obtained based on the equivalent circuit of the multi-source partial discharge detection device in the high-voltage switch cabinet, C1 is the equivalent capacitance between the partial discharge source and the needle electrode in the equivalent circuit, C2 is the equivalent capacitance between the needle electrode and the plate electrode, R is the equivalent resistance of the multi-source partial discharge detection device in the high-voltage switch cabinet, s is a parameter introduced in the Laplace change, where s=jω.
[0015] Preferably, in the broadband sensor:
[0016] The height of the annular protrusion on the insulating layer plate is not less than the thickness of the plate electrode;
[0017] The needle electrode and the insulating layer plate are positioned perpendicular to each other;
[0018] The side of the circuit board is fixedly connected to the lower surface of the plate electrode, and the needle electrode is connected to the circuit board through a needle electrode connector. The needle electrode connector has a corner and is a solid structure. The plate electrode connector is coaxially arranged with the needle electrode connector, and the distance between the inner wall of the plate electrode connector and the needle electrode connector is d.
[0019] Preferably, the acquisition card is a four-channel data acquisition card with a sampling rate of 1.25 GS / s and a sampling bit number of 8 bits.
[0020] Preferably, the implementation subject is a detection system composed of the aforementioned multi-source partial discharge detection device in the high-voltage switch cabinet and a computer terminal, and the steps include:
[0021] Step S1, in the switch cabinet group to be tested, a switch cabinet located at the end is selected as the master end sensor measuring point, and a switch cabinet with a distance D from the master end sensor measuring point is selected as the slave end sensor measuring point, and the detection device is started;
[0022] Step S2, the detection device performs synchronous sampling according to the sampling frequency and sampling depth preset by the computer terminal to obtain a mixed partial discharge sampling signal group X from the master end sensor m’ , and a mixed partial discharge sampling signal group X from the slave sensor s’ ;
[0023] Step S3: the detection device detects the X m’ and the X s’ Filter and reduce noise, extract the partial discharge pulse waveform of each signal, and group the obtained mixed partial discharge pulse signal of the master end sensor into X m , and the mixed partial discharge pulse signal group X of the slave end sensor s Output to the computer terminal, where Xm ={X m1 ,X m2 ,…X mi};X s ={X s1 ,X s2 ,…X sj}, pulse signal X mi is a pulse signal corresponding to a partial discharge signal of a partial discharge source collected by the master-end sensor, and the pulse signal X sj A pulse signal corresponding to a partial discharge signal of a certain partial discharge source collected by the slave end sensor;
[0024] Step S4, the computer terminal receives the mixed partial discharge pulse signal group X sent by the detection device m and the mixed partial discharge pulse signal group X s ;
[0025] Step S5: m All pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain a set of k pulse signals, which is recorded as Y mk , Y mk = {Y mk1 ,Y mk2 ,…,Y mkp}, the set Y mk The elements in are arranged in the order of acquisition time; for the signal group X s All pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain n groups of pulse signals, which are recorded as Y sn , Y sn = {Y sn1 ,Y sn2 ,…,Y snq}, the set Y sn The elements in are arranged in the order of acquisition time;
[0026] Step S6: for each set Y mk With each of the sets Y sn Perform paired analysis in sequence, so that each of the sets Y mk There is a corresponding set Y sn , at this time p = q, the Y mk and the Y sn There are p elements in each case, and step S6 generates k pairs of pairs in total;
[0027] Step S7, arranging the set Y in order mk With the set Y snThe elements in are paired one by one, and p groups of arrival time point data are obtained by using the time window and calculating the correlation coefficient. The arrival time point data includes the time point t at which the partial discharge source arrives at the main end sensor measuring point. p1 And the time point t of the slave sensor measurement point p2 ;
[0028] Step S8, determining p positioning positions of the same partial discharge source according to the p groups of arrival time point data and the propagation speed V of the pulse signal;
[0029] Step S9, according to the k pairs of pairings obtained in step S6, based on step S8, obtain the positioning distribution map of the k partial discharge sources.
[0030] Preferably, the step S5 performs the following steps on the signal group X: m The steps of classifying and grouping partial discharge sources and filtering out pulse noise signals for all pulse signals in the system include:
[0031] Step S51, calculate the signal group X in sequence m The mutual correlation coefficient a between the two signals x ;
[0032] Step S52, satisfying a x The two signals with the condition of ≥a1 are classified into the same group, and a total of k groups are obtained to obtain the pulse signal set Y of k groups mk , where a is not satisfied with any other signal x The signal ≥a1 is identified as a pulse noise signal and is selected from the signal group X m Removed;
[0033] Step S53, arranging the pulse signal set Y in the order of acquisition time mk The elements in .
[0034] Preferably, the step S6 performs the following operations on each set Y: mk With each of the sets Y sn The steps to perform a sequential paired analysis include:
[0035] Step S61, select a set Y mk and a set Y sn , take the first element Y of the set mk1 and Y sn1 Calculate the mutual correlation coefficient a between two signals Y ;
[0036] Step S62, satisfying a Y Pair two sets ≥a2 to obtain k pairs of pairing results.
[0037] Preferably, the step S7 comprises:
[0038] Step S71, select the set Y mk The first element Y in mk1 With the set Y sn The element Y at the corresponding position sn1 The pairing is group 1;
[0039] Step S72, find the Y mk1 The peak time is t 11 , taking the peak moment as the center, select the Y mk1 The length is t(Y mk1 ) / 2 time window Z, t(Y mk1 ) is the Y mk1 Pulse signal length;
[0040] Step S73, from Y sn1 Start from the beginning of t(Y mk1 ) / 2 time window Z s , calculate the pulse signal in time window Z and time window Z s The mutual correlation coefficient a between the inner pulse signals Z1 , slide the time window Z backward s , after each sliding, the mutual correlation coefficient a is calculated Zx , we get a set of mutual correlation numbers {a Z1 ,a Z12 ,…,a Zx}, select the time window Z corresponding to the maximum value in the set of correlation coefficients s The time difference △t between the time window Z and Y sn1 The arrival time is t 12 , where t 12 =t 11 +△t,t 11 and t 12 This is the arrival time point data of the first group;
[0041] Step S74, select the set Y again mk The second element Y in mk2 With the set Y sn The element Y at the corresponding position sn2 Pairing is performed into the second group and the second group of arrival time point data is calculated, and pairing is performed in sequence until the arrival time point data of the p group is calculated.
[0042] Preferably, the step S8 determines p locations of the same partial discharge source according to the p groups of arrival time point data and the propagation speed V of the pulse signal, including:
[0043] Step S81, according to the time point t in a set of arrival time point data p1 , the time point t p2 and the propagation speed V of the pulse signal, and calculate the distance between the partial discharge source and the main end sensor measuring point;
[0044] Step S82, calculating p positioning positions of the same partial discharge source according to p groups of arrival time point data.
[0045] It can be seen from the above technical solution that the multi-source partial discharge detection device and positioning method in the high-voltage switch cabinet provided by the embodiment of the present invention include the following steps: step S1, in the switch cabinet group to be tested, a switch cabinet located at the end is selected to be set as the main end sensor measuring point, and a switch cabinet with a distance of D from the main end sensor measuring point is selected to be set as the slave end sensor measuring point, and the detection device is started; step S2, the detection device performs synchronous sampling according to the sampling frequency and sampling depth preset by the computer terminal, and obtains a mixed partial discharge sampling signal group X from the main end sensor m’ , and the mixed partial discharge sampling signal group X from the slave sensor s’ Step S3, the detection device detects X m’ and X s’ Filter and reduce noise, extract the partial discharge pulse waveform of each signal, and group the mixed partial discharge pulse signal of the main end sensor into X m , and the mixed partial discharge pulse signal group X of the slave sensor s Output to the computer terminal; Step S4, the computer terminal receives the mixed partial discharge pulse signal group X sent by the detection device m And mixed partial discharge pulse signal group X s ; Step S5, for signal group X m All pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain a set of k pulse signals, which is recorded as Y mk , Y mk = {Y mk1 ,Y mk2 ,…,Y mkp}, set Y mk The elements in are arranged in the order of acquisition time; for the signal group X s All pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain n groups of pulse signals, which are recorded as Y sn , Y sn = {Y sn1 ,Y sn2 ,…,Y snq}, set Y sn The elements in are arranged in the order of collection time; Step S6, for each set Y mk With each set Y sn Perform paired analysis in sequence so that each set Ymk There is a corresponding set Y sn , at this time p=q, Y mk and Y sn There are p elements in each set, and step S6 generates k pairs of pairs in total; step S7, sort the set Y in the order of arrangement mk With the set Y sn The elements in are paired one by one, and the time window and the correlation coefficient are calculated to obtain p groups of arrival time point data. The arrival time point data includes the time point t when the partial discharge source arrives at the main end sensor measurement point. p1 And the time point t of the slave sensor measurement point p2 ; Step S8, according to the p groups of arrival time point data and the propagation speed V of the pulse signal, p positioning positions of the same partial discharge source are determined; Step S9, according to the k pairs of pairs obtained in step S6, the positioning distribution map of k partial discharge sources is obtained based on step S8. The method of the present invention can realize the accurate positioning of multiple partial discharge sources based on multi-source partial discharge mixed signals. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1 The figure is a schematic diagram of the working principle of the multi-source partial discharge detection device in the high-voltage switch cabinet of the present invention.
[0047] Figure 2 FIG. 4 is a structural diagram of a broadband sensor of the present invention.
[0048] Figure 3 It is the equivalent schematic diagram of the multi-source partial discharge detection device in the high-voltage switchgear.
[0049] Figure 4 It is a wiring schematic diagram of a circuit board in a broadband sensor of the present invention;
[0050] Figure 5 This is a bandwidth comparison diagram of the sensor of the multi-source partial discharge detection device in the high-voltage switch cabinet of the present invention, TEV and HFCT.
[0051] Figure 6 The present invention is a flow chart of a method for detecting multi-source partial discharge in a high-voltage switch cabinet.
[0052] Figure 7 The present invention is a schematic diagram of the operation of the multi-source partial discharge detection device in the high-voltage switch cabinet to detect the switch cabinet.
[0053] Figure 8 It is a diagram showing the application effect of the device of the present invention.
[0054] In the figure: a master end sensor 1, a slave end sensor 2, a microprocessor 3, a high frequency data acquisition unit 4, a filter amplification unit 5, a power supply unit 6, a needle electrode 11, an insulating layer plate 12, a plate electrode 13, a circuit board 14, and a plate electrode connector 15. DETAILED DESCRIPTION
[0055] The technical scheme and technical effects of the present invention are further elaborated in detail below in conjunction with the accompanying drawings of the present invention.
[0056] The present invention provides a multi-source partial discharge detection device in a high-voltage switch cabinet, which can detect and preliminarily locate the partial discharge that has occurred inside the switch cabinet during normal operation. Figure 1-Figure 4 As shown, the multi-source partial discharge detection device in the high-voltage switchgear is composed of a broadband sensor and a host, wherein the broadband sensor is a high-precision, large-bandwidth partial discharge sensor component composed of a needle electrode and a metal plate electrode, and each sensor is based on the principle of an electrically small antenna, and completes the function of collecting partial discharge signals by collecting the potential difference in the electric field of the control panel. Figure 1 In the invention, two broadband sensors are used as the master sensor 1 and the slave sensor 2 respectively, and the host is composed of a microprocessor 3, a high-frequency data acquisition unit 4, a filter amplifier unit 5 and a power supply unit 6;
[0057] like Figure 2As shown, the broadband sensor includes a needle electrode 11, an insulating layer plate 12, a circular plate electrode 13 and a circuit board 14 which are arranged in sequence from top to bottom: a through hole matching the size of the needle electrode 11 is provided at the center of the insulating layer plate, which is located on the lower surface of the insulating layer plate 12, and the edge of the through hole has an annular protrusion; the needle electrode 11 is inserted into the through hole, the needle electrode is fixedly connected to the inner surface of the through hole, the positional relationship between the needle electrode and the insulating layer plate is mutually perpendicular, the head of the needle electrode 11 is used to sense an electric field signal, and the end of the needle electrode 11 is connected to the positive input terminal in+ of the circuit board 14; a through hole is provided at the center of the plate electrode 13, the upper surface of the plate electrode 13 is fixedly connected to the lower surface of the insulating layer plate 12, the annular protrusion of the insulating layer plate 12 passes through the through hole, and the height of the annular protrusion on the insulating layer plate is not less than the thickness of the plate electrode; the plate electrode 13 is used to sense an electric field signal, and a hollow plate is passed between the plate electrode 13 and the circuit board 14. The electrode connector 15 is fixedly connected, and the plate electrode connector 15 is electrically connected to the negative input terminal in- of the circuit board 14 and the ground terminal GND; the needle electrode 11 and the plate electrode connector 13 are coaxially arranged, and the distance between the needle electrode and the inner wall of the plate electrode connector is d, the needle electrode 11 serves as an inner conductor, the plate electrode 13 and the plate electrode connector 15 serve as outer conductors, the insulating layer plate 12 serves as an insulating medium, the needle electrode and the plate electrode together constitute the two poles of the antenna, and are isolated by the insulating layer plate 12, the needle electrode is connected to the coaxial inner core, and is connected to the positive input terminal of the conditioning circuit, the plate electrode is connected to the negative input terminal of the conditioning circuit through the coaxial outer shielding layer, and the three are combined to form a coaxial feeding form; the circuit board 14 is powered by a lithium battery, the circuit board 14 has a positive input terminal, a negative input terminal, a ground terminal and an output terminal out, the output terminal of the circuit board 14 is connected to the host through a coaxial transmission line, and the output signal U of the circuit board 14 is connected to the host through a coaxial transmission line. out is the needle electrode sensing signal U after conditioning and amplification N The plate electrode sensing signal U P The result is the signal difference between .
[0058] The needle electrode 11 can be directly connected to the circuit board 14. In this case, the circuit board 14 and the plate electrode 13 are arranged in parallel. The plate electrode connector 15 can be a cylindrical structure. The needle electrode 11 is located at the center of the cylindrical structure. The needle electrode and the inner wall of the cylindrical structure maintain a distance d.
[0059] Another form, such as Figure 2As shown, the side of the circuit board 14 is fixedly connected to the lower surface of the plate electrode 13, and the needle electrode 11 is connected to the circuit board 14 through the needle electrode connector 16. Accordingly, the plate electrode connector 15 needs to have a corner structure, and the plate electrode connector is coaxially arranged with the needle electrode connector 16, and the distance between its inner wall and the needle electrode connector 16 is d. The close connection between the circuit board 14 and the lower surface of the plate electrode 13 can avoid the refraction and reflection of the signal during the coaxial transmission process, and effectively improve the signal-to-noise ratio of the sensor.
[0060] From the above, it can be seen that the broadband sensor is a broadband wireless needle plate sensor based on the principle of small electric antenna, which contains conditioning amplification and filtering circuits to measure partial discharge signals. Since the size of the needle plate sensor is much smaller than the wavelength of the partial discharge pulse, its working principle is equivalent to a capacitor in an electrostatic field. There are also capacitance values between the needle electrode and the partial discharge source, and between the plate electrode and the ground. Therefore, the broadband sensor senses the voltage jump caused by partial discharge inside the switch cabinet through the principle of capacitive voltage divider, and senses the signal.
[0061] The high-frequency data acquisition unit 4 is a four-channel data acquisition card with a sampling rate of up to 1.25GS / s and a sampling bit of 8 bits. It can connect multiple sensors at the same time for synchronized triggering and data acquisition;
[0062] The filter amplifier unit 5 is a single-channel filter amplifier circuit that receives the signal of the broadband sensor and performs filtering and noise reduction on it. Figure 4 As shown, the single-channel filter amplifier circuit is composed of an ADA4817 chip and a bridge filter circuit. The high-pass filter composed of resistors and capacitors can adjust the bandwidth of the sensor so that its fluctuation between 1-200MHz does not exceed 6dB. The amplifier circuit is a common-phase proportional amplifier circuit composed of ADA4817-1, and its signal gain is G=10. The working principle of the broadband sensor in the electrostatic field of the present invention is equivalent to a capacitor. For capacitive sensors, the capacitance of the sensor is usually very small, and in actual use, there is the influence of stray capacitance. This makes the change of the surrounding environment cause the change of stray capacitance, thereby affecting the sensitivity of the sensor and interfering with its bandwidth. Therefore, for capacitive sensors, the bridge method is usually used as a conditioning filter circuit. Figure 4 The feed output of a broadband sensor is shown connected to a resistor-capacitor bridge and isolated by an operational amplifier. In this case, the sensor can operate in a balanced state, remain stable in the frequency band of 1-200MHz, and achieve a gain of 10.
[0063] The microprocessor 3 is connected to the high-frequency data acquisition unit 4 and the filter amplifier unit 5, and is used to trigger the high-frequency data acquisition unit 4 to collect signals; receive the signal processed by the filter amplifier unit 5 and send it to the computer terminal. The microprocessor 3 is a central control processor with FPGA as the core, which is responsible for controlling the triggering, collection, communication and other functions of the entire system. When the user issues an instruction to the microprocessor through the computer, the processor triggers each channel of the acquisition card in a very short time, realizes the synchronous collection of the signal, and filters and amplifies the collected signal through the conditioning circuit. Then the pulse signal in the data is extracted, and the arrival time of each pulse signal collected by each channel is recorded one by one with the trigger event of the acquisition card as the time zero point. The microprocessor 3 can be used to control the output of the power supply unit 6, ensure that the output current is sufficient and reduce the voltage ripple; provide a USB Hub interface, which can realize the communication between the computer and the control unit, and can also complete the data transmission with the acquisition card; provide a trigger signal, send a signal to the data acquisition card through a coaxial transmission line, and realize the triggering of the acquisition card.
[0064] The power supply unit 6 is composed of a lithium battery and a voltage stabilizing circuit, and its function is to power the entire instrument. The voltage stabilizing circuit first steps down the voltage through the LTM8045, and then achieves a regulated output through the LDO. While ensuring the output power, the ripple level of the power supply output voltage is greatly reduced, thereby reducing the system noise and improving the detection sensitivity. Specifically, it can be a 21V lithium battery pack.
[0065] like Figure 3 As shown in the figure, it is the equivalent circuit diagram of the multi-source partial discharge detection device in the high-voltage switch cabinet when it is working. Since the size of the broadband sensor is much smaller than the electromagnetic wave wavelength of the partial discharge signal, the surrounding environment can be regarded as an electrostatic field. Where C1 is the equivalent capacitance between the partial discharge source and the needle electrode, C2 is the equivalent capacitance between the sensor needle plate electrodes, R is the equivalent resistance of the multi-source partial discharge detection device in the high-voltage switch cabinet, and s is the parameter introduced in the Laplace change, where s = jω. The output signal U of the broadband sensor out The transfer function of the partial discharge signal U in the switch cabinet is shown in Formula 1:
[0066]
[0067] like Figure 5 As shown in the figure, the amplitude-frequency / phase-frequency characteristic curves of the multi-source partial discharge detection device in the high-voltage switch cabinet of the present invention and the traditional HFCT and TEV sensors are shown. It can be seen from the figure that the wideband sensor of the present invention can significantly increase the detection bandwidth compared with HFCT, and the non-contact sensing method is also more convenient in actual use. The wideband sensor and the TEV sensor are both capacitive sensors, and their working principles are roughly the same, but the sensitivity has been greatly improved.
[0068] In the present invention, the high-frequency data acquisition unit 4 can be connected to multiple sensors at the same time for synchronized triggering and data acquisition, and can be connected to multiple slave sensors for signal acquisition, which is beneficial to improving the positioning accuracy of partial discharge.
[0069] Furthermore, if Figure 6 As shown, the present invention also provides a method for detecting multi-source partial discharge in a high-voltage switch cabinet, the implementation subject includes Figure 1-4 The multi-source partial discharge detection device and computer terminal in the high-voltage switch cabinet shown in the figure include the following steps:
[0070] Step S1, in the switch cabinet group to be tested, a switch cabinet located at the end is selected as the master end sensor measuring point, and a switch cabinet with a distance D from the master end sensor measuring point is selected as the slave end sensor measuring point, and the detection device is started;
[0071] Step S2: The detection device performs synchronous sampling according to the sampling frequency and sampling depth preset by the computer terminal to obtain a mixed partial discharge sampling signal group X from the master end sensor 1. m’ , and the mixed partial discharge sampling signal group X from the slave sensor 2 s’ ;
[0072] Step S3: the detection device detects X m’ and X s’ Filter and reduce noise, extract the partial discharge pulse waveform of each signal, and group the mixed partial discharge pulse signal of the main end sensor into X m , and the mixed partial discharge pulse signal group X of the slave sensor s Output to the computer terminal, where X m ={X m1 ,X m2 ,…X mi};X s ={X s1 ,X s2 ,…X sj}, pulse signal X mi It is the pulse signal corresponding to the partial discharge signal of a certain partial discharge source collected by the main end sensor. The pulse signal X sj It is a pulse signal corresponding to the partial discharge signal of a certain partial discharge source collected by the end sensor;
[0073] Step S4: The computer terminal receives the mixed partial discharge pulse signal group X sent by the detection device. m And mixed partial discharge pulse signal group X s ;
[0074] Step S5: mAll pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain a set of k pulse signals, which is recorded as Y mk , Y mk = {Y mk1 ,Y mk2 ,…,Y mkp}, set Y mk The elements in are arranged in the order of acquisition time; for the signal group X s All pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain n groups of pulse signals, which are recorded as Y sn , Y sn = {Y sn1 ,Y sn2 ,…,Y snq}, set Y sn The elements in are arranged in the order of acquisition time. Here, the actual number of groups n is the same as the number of groups k, both representing the total number of current partial discharge sources.
[0075] Step S6: for each set Y mk With each set Y sn Perform paired analysis in sequence so that each set Y mk There is a corresponding set Y sn , at this time p=q, Y mk and Y sn There are p elements in each case, and step S6 generates a total of k pairs of pairs;
[0076] Step S7: sort the set Y in order mk With the set Y sn The elements in are paired one by one, and the time window and the correlation coefficient are calculated to obtain p groups of arrival time point data. The arrival time point data includes the time point t when the partial discharge source arrives at the main end sensor measurement point. p1 And the time point t of the slave sensor measurement point p2 ;
[0077] Step S8, determining p locations of the same partial discharge source according to p groups of arrival time point data and the propagation speed V of the pulse signal;
[0078] Step S9, according to the k pairs of pairs obtained in step S6, based on step S8, obtain the positioning distribution map of k partial discharge sources.
[0079] Among them, step S5 is for signal group X m The specific implementation steps for classifying and grouping partial discharge sources and filtering out pulse noise signals for all pulse signals in the system include:
[0080] Step S51, calculate the signal group X in sequence m The mutual correlation coefficient a between the two signals x;
[0081] Step S52, satisfying a x The two signals with the condition ≥a1 are classified into the same group, and a total of k groups are obtained, and the k-group pulse signal set Y mk , where a is not satisfied with any other signal x The signal ≥a1 is identified as a pulse noise signal and is selected from the signal group X m Removed;
[0082] Step S53: Arrange the pulse signal set Y in the order of acquisition time mk The elements in .
[0083] Among them, step S6 for each set Y mk With each set Y sn The specific implementation steps for sequential paired analysis include:
[0084] Step S61, select a set Y mk and a set Y sn , take the first element Y of the set mk1 and Y sn1 Calculate the mutual correlation coefficient a between two signals Y ;
[0085] Step S62, satisfying a Y Pair two sets ≥a2 to obtain k pairs of pairing results.
[0086] The specific operation of locating the partial discharge source in step S7 is as follows:
[0087] Step S71, select set Y mk The first element Y in mk1 With the set Y sn The element Y at the corresponding position sn1 The pairing is group 1;
[0088] Step S72, find Y mk1 The peak time is t 11 , taking the peak moment as the center, select Y mk1 The length is t(Y mk1 ) / 2 time window Z, t(Y mk1 ) is Y mk1 Pulse signal length;
[0089] Step S73, from the Y sn1 Start from the beginning of t(Y mk1 ) / 2 time window Z s , calculate the pulse signal in the time window Z and the time window Z s The mutual correlation coefficient a between the inner pulse signalsZ1 , slide the time window Z backward s , after each sliding, the mutual correlation coefficient a is calculated Zx , we get a set of mutual correlation numbers {a Z1 ,a Z12 ,…,a Zx}, select the time window Z corresponding to the maximum value in the set of correlation coefficients s The time difference △t between the time window Z and the Y sn1 The arrival time is t 12 , where t 12 =t 11 +△t, t 11 and the 12 This is the arrival time point data of the first group;
[0090] Step S74, select set Y again mk The second element Y in mk2 With the set Y sn The element Y at the corresponding position sn2 Pair them into the second group and calculate the arrival time point data of the second group. Pair them in sequence until the arrival time point data of the p groups are calculated.
[0091] Step S8 determines the p locations of the same partial discharge source according to the p groups of arrival time point data and the propagation speed V of the pulse signal. The specific implementation includes:
[0092] Step S81, according to a set of arrival time point data, the time point t p1 , time point t p2 And the propagation speed V of the pulse signal, calculate the distance between the partial discharge source and the main end sensor measurement point;
[0093] Step S82, calculating p positioning positions of the same partial discharge source according to p groups of arrival time point data.
[0094] When the microprocessor does not receive the trigger acquisition command, it always maintains the synchronous trigger acceptance mode. When the computer issues a synchronous trigger command to the processor, the processor configures the trigger mode, sampling rate, acquisition depth, etc. of the acquisition card, and injects a synchronous pulse trigger signal into the external trigger channel of the A / D conversion acquisition card after a delay of 1s. When the external trigger channel of the acquisition card receives the synchronous pulse, the synchronous pulse triggers the detection data of the high-precision wireless sensor of the A / D conversion acquisition card through the pulse conditioning circuit, and transmits the detection data to the microprocessor. The microprocessor first reduces the noise of the collected signal through the noise reduction algorithm, extracts the corresponding partial discharge pulse waveform, and records the arrival time of each waveform. The processor sends the extracted pulse data to the computer via USB, which is analyzed by the computer.
[0095] like Figure 7 As shown, it is a field embodiment of the present invention. The broadband sensor of the present invention is installed on the outer surface of both ends of a row of multi-faceted high-voltage switch cabinets, which can effectively detect the partial discharge signal generated by the high-voltage switch cabinet. The host can be placed near the switch cabinet, and the two broadband sensors are connected to the two detection channels of the detection host through a coaxial transmission line. When the microprocessor receives the synchronous acquisition signal, the A / D acquisition card starts to collect the partial discharge signals sensed by the two sensors at high speed, and uploads the extracted partial discharge pulse data to the computer, which can effectively realize the detection of partial discharge signals of multi-faceted switch cabinets and quickly find the location of multiple partial discharge sources inside. For the staff, through the positioning spectrum displayed on the computer after the detection, it is possible to determine the switch cabinet with partial discharge, and then realize the work of eliminating the fault inside the switch cabinet. The detected phase spectrum (PRPD) can roughly realize the pattern recognition of partial discharge.
[0096] like Figure 8 The figure is a schematic diagram of the working effect of a multi-source partial discharge detection and positioning device for a switch cabinet. The high-precision wireless sensors placed on both sides of the switch cabinet collect partial discharge signals in several power frequency cycles. The time domain waveform after conditioning and amplification is as follows: Figure 7As shown in the two figures above. It can be seen from the figure that the collected signal mainly includes three parts: partial discharge pulse signal in various forms, white noise signal, and pulse noise signal. When the sensor completes the signal collection, the system will first reduce the noise of the collected signal through the corresponding filtering and noise reduction algorithm, so as to remove the white noise in the system and the surrounding environment. Compared with the traditional single-ended detection, the characteristics of the pulse noise in waveform, amplitude, and phase are not much different from those of the typical partial discharge. The traditional partial discharge detection technology will misjudge this noise as partial discharge, which also makes the switch cabinet under a complex electromagnetic environment. The detection of partial discharge is not reliable enough. However, the pulse noise does not have the characteristic of multiple repetitions. Under the detection environment of the invention, the signals received by the sensors at the left and right ends will repeatedly extract pulses and obtain the mutual correlation coefficients between multiple pulses. The mutual correlation between the pulse noise and other pulse signals is extremely low. This method can eliminate such noise and only retain the partial discharge pulse signal. In addition, for the partial discharge source, due to the different types of defects and the differences in circuit topology, the pulse waveforms generated by different partial discharge sources when partial discharge occurs are also different. Therefore, multiple partial discharge pulses generated by the same partial discharge source have great correlation, while the correlation between partial discharge pulses from different partial discharge sources is small. By classifying partial discharges by checking the cross-correlation between pulse waveforms, separation of multi-source partial discharges can also be achieved. When determining the arrival time of the signal, the positioning accuracy of the partial discharge can be greatly improved by obtaining the cross-correlation coefficient and finding its maximum point as the arrival time of the signal. It can be seen from the positioning spectrum that there will be multiple positioning marks for the same partial discharge source, and the three clusters of positioning results gathered at different positions successfully separated and located the multi-source partial discharges. For the signal generated by the same partial discharge source, the present invention can also extract it separately, check its waveform information and PRPD spectrum, which is helpful for pattern recognition of partial discharge.
[0097] The present invention uses a cross-correlation algorithm to process partial discharge signals. If there are multiple partial discharge sources in the switch cabinet, the partial discharge signal collected by the same sensor should include the partial discharge signal generated by each partial discharge source at the same time. There are also multiple partial discharge signals generated by each partial discharge source. First, the pulse signal is extracted in an array through the rising and falling edges of the pulse signal, and the correlation between all pulses is calculated. Since the similarity of the multiple pulse waveforms generated by the same partial discharge source is extremely large, the cross-correlation coefficient is close to 1. The pulses generated by different partial discharge sources have large waveform differences and small cross-correlation coefficients. In addition, for the same partial discharge generated by the same partial discharge source, the sensors at both ends of the switch cabinet will collect very similar pulse waveforms. By establishing a time window of the same length as the pulse waveform and continuously sliding the window, the cross-correlation coefficient of the two waveforms is calculated, and the time point with the largest cross-correlation coefficient is found as the arrival time of the partial discharge signal, and then the position is calculated according to the time difference. While effectively reducing the influence of waveform distortion caused by high-frequency attenuation and dispersion effects, this method avoids positioning errors when there are multiple partial discharge sources in the switch cabinet, and greatly improves the positioning accuracy of partial discharge.
[0098] The multi-source partial discharge detection device in the high-voltage switch cabinet provided by the present invention realizes the detection and preliminary positioning of the partial discharge that has occurred inside the switch cabinet under normal operation. The broadband sensor used has the characteristics of broadband and high sensitivity, and can synchronously collect partial discharge signals within the range of 1-200HMz. Compared with the traditional UHF or TEV, the sensor can realize high-precision and large-bandwidth detection of partial discharge in a non-contact situation; in the case where there are multiple partial discharge sources inside the switch cabinet, the partial discharge pulses are classified and the arrival time of the signal is determined by obtaining the mutual correlation coefficient between the pulse waveforms, so as to realize the separation of multi-source partial discharges and greatly improve the accuracy of partial discharge positioning, solving the problem of difficulty in positioning partial discharge when there are multiple discharge sources in the current switch cabinet. The multi-source partial discharge detection device in the high-voltage switch cabinet provided by the present invention is simple to operate, easy to use, highly reliable, easy to carry, and provides convenience for the daily inspection of partial discharge inside the switch cabinet.
[0099] What is disclosed above is only a preferred embodiment of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. A person skilled in the art can understand that all or part of the processes of the above embodiments and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.
Claims
1. A multi-source partial discharge detection device in a high-voltage switch cabinet, characterized in that: include: Two broadband sensors and a host, wherein the two broadband sensors serve as a master sensor and a slave sensor respectively, and the host is composed of a microprocessor, a high-frequency data acquisition unit, a filter amplifier unit, and a power supply unit; The broadband sensor comprises a needle electrode, an insulating layer plate, a circular plate electrode and a circuit board which are arranged in sequence from top to bottom: a through hole matching the size of the needle electrode is provided in the center of the insulating layer plate and is located on the lower surface of the insulating layer plate, and the edge of the through hole has an annular protrusion; the needle electrode is inserted into the through hole, the needle electrode is fixedly connected to the inner surface of the through hole, the head of the needle electrode is used to sense an electric field signal, and the end of the needle electrode is connected to the positive input terminal of the circuit board; a through hole is provided in the center of the plate electrode, the upper surface of the plate electrode is fixedly connected to the lower surface of the insulating layer plate, and the annular protrusion of the insulating layer plate passes through the through hole; the plate electrode is used to sense an electric field signal, The plate electrode and the circuit board are fixedly connected via a hollow plate electrode connector, and the plate electrode connector is electrically connected to the negative input terminal and the ground terminal of the circuit board; the needle electrode and the plate electrode connector are coaxially arranged, and the distance between the needle electrode and the inner wall of the plate electrode connector is d. The needle electrode serves as an inner conductor, the plate electrode and the plate electrode connector serve as an outer conductor, and the insulating layer plate serves as an insulating medium, and the three are combined to form a coaxial feeding form; the circuit board is powered by a battery, and the circuit board has the positive input terminal, the negative input terminal, the ground terminal and the output terminal, and the output terminal of the circuit board is connected to the host through a coaxial transmission line, and the output signal U of the circuit board out is the needle electrode sensing signal U after conditioning and amplification N The plate electrode sensing signal U P The high-frequency data acquisition unit is used to collect the output signals of the two broadband sensors through an acquisition card; the filtering and amplifying unit is used to filter and reduce noise on the output signals of the two broadband sensors; The microprocessor is connected to the high-frequency data acquisition unit and the filter amplification unit, and is used to trigger the high-frequency data acquisition unit to collect signals; receive the signals processed by the filter amplification unit and send them to the computer terminal; The power supply unit is used to supply power to the microprocessor, the high-frequency data acquisition unit and the filter amplification unit; The broadband sensor collects partial discharge signals in a frequency band ranging from 1 MHz to 200 MHz. The output signal U out And the partial discharge signal U in the switch cabinet sig The corresponding relationship is: The above formula is obtained based on the equivalent circuit of the multi-source partial discharge detection device in the high-voltage switch cabinet, C1 is the equivalent capacitance between the partial discharge source and the needle electrode in the equivalent circuit, C2 is the equivalent capacitance between the needle electrode and the plate electrode, R is the equivalent resistance of the multi-source partial discharge detection device in the high-voltage switch cabinet, s is a parameter introduced in the Laplace change, where s=jω.
2. The multi-source partial discharge detection device in a high-voltage switch cabinet according to claim 1, characterized in that: In the broadband sensor: The height of the annular protrusion on the insulating layer plate is not less than the thickness of the plate electrode; The positional relationship between the needle electrode and the insulating layer plate is perpendicular to each other; The side of the circuit board is fixedly connected to the lower surface of the plate electrode, and the needle electrode is connected to the circuit board through a needle electrode connector. The needle electrode connector has a corner and is a solid structure. The plate electrode connector is coaxially arranged with the needle electrode connector, and the distance between the inner wall of the plate electrode connector and the needle electrode connector is d.
3. The multi-source partial discharge detection device in a high-voltage switch cabinet according to claim 2, characterized in that: The acquisition card is a four-channel data acquisition card with a sampling rate of 1.25 GS / s and a sampling bit number of 8 bits.
4. A method for locating multi-source partial discharge in a high-voltage switch cabinet, characterized in that: The implementation subject is a detection system composed of a multi-source partial discharge detection device in a high-voltage switch cabinet and a computer terminal as described in any one of claims 1 to 3, and the steps include: Step S1, in the switch cabinet group to be tested, a switch cabinet located at the end is selected as the master end sensor measuring point, and a switch cabinet with a distance D from the master end sensor measuring point is selected as the slave end sensor measuring point, and the detection device is started; Step S2, the detection device performs synchronous sampling according to the sampling frequency and sampling depth preset by the computer terminal to obtain a mixed partial discharge sampling signal group X from the master end sensor m’ , and a mixed partial discharge sampling signal group X from the slave end sensor s’ ; Step S3: the detection device detects the X m’ and the X s’ Filter and reduce noise, extract the partial discharge pulse waveform of each signal, and group the mixed partial discharge pulse signal of the master end sensor into X m , and the mixed partial discharge pulse signal group X of the slave end sensor s Output to the computer terminal, where X m ={X m1 ,X m2 ,…X mi };X s ={X s1 ,X s2 ,…X sj }, pulse signal X mi is a pulse signal corresponding to a partial discharge signal of a partial discharge source collected by the master-end sensor, and the pulse signal X sj A pulse signal corresponding to a partial discharge signal of a certain partial discharge source collected by the slave end sensor; Step S4, the computer terminal receives the mixed partial discharge pulse signal group X sent by the detection device m and the mixed partial discharge pulse signal group X s ; Step S5: m All pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain a set of k pulse signals, which is recorded as Y mk , Y mk = {Y mk1 ,Y mk2 ,…,Y mkp }, the set Y mk The elements in are arranged in the order of acquisition time; for the signal group X s All pulse signals in the PD source are classified and grouped, and pulse noise signals are screened out to obtain n groups of pulse signals, which are recorded as Y sn , Y sn = {Y sn1 ,Y sn2 ,…,Y snq }, the set Y sn The elements in are arranged in the order of acquisition time; Step S6: for each set Y mk With each of the sets Y sn Perform paired analysis in sequence, so that each of the sets Y mk There is a corresponding set Y sn , at this time p = q, the Y mk and the Y sn There are p elements in each case, and step S6 generates k pairs of pairs in total; Step S7, arranging the set Y in order mk With the set Y sn The elements in are paired one by one, and p groups of arrival time point data are obtained by using the time window and calculating the correlation coefficient. The arrival time point data includes the time point t at which the partial discharge source arrives at the main end sensor measuring point. p1 And the time point t of the slave sensor measurement point p2 ; Step S8, determining p positioning positions of the same partial discharge source according to the p groups of arrival time point data and the propagation speed V of the pulse signal; Step S9, according to the k pairs of pairings obtained in step S6, based on step S8, obtain the positioning distribution map of the k partial discharge sources.
5. The method for locating multi-source partial discharge in a high-voltage switch cabinet according to claim 4, characterized in that: The step S5 performs a m The steps of classifying and grouping partial discharge sources and filtering out pulse noise signals for all pulse signals in the system include: Step S51, calculate the signal group X in sequence m The mutual correlation coefficient a between the two signals x ; Step S52, satisfying a x The two signals with the condition of ≥a1 are classified into the same group, and a total of k groups are obtained to obtain the pulse signal set Y of k groups mk , where a is not satisfied with any other signal x The signal ≥a1 is identified as a pulse noise signal and is selected from the signal group X m Removed; Step S53, arranging the pulse signal set Y in the order of acquisition time mk The elements in .
6. The method for locating multi-source partial discharge in a high-voltage switch cabinet according to claim 5, characterized in that: The step S6 performs a mk With each of the sets Y sn The steps to perform a sequential paired analysis include: Step S61, select a set Y mk and a set Y sn , take the first element Y of the set mk1 and Y sn1 Calculate the mutual correlation coefficient a between two signals Y ; Step S62, satisfying a Y Pair two sets ≥a2 to obtain k pairs of pairing results.
7. The method for locating multi-source partial discharge in a high-voltage switch cabinet according to claim 6, characterized in that: The step S7 comprises: Step S71, select the set Y mk The first element Y in mk1 With the set Y sn The element Y at the corresponding position sn1 The pairing is group 1; Step S72, find the Y mk1 The peak time is t 11 , taking the peak moment as the center, select the Y mk1 The length is t(Y mk1 ) / 2 time window Z, t(Y mk1 ) is the Y mk1 Pulse signal length; Step S73, from the Y sn1 Start from the beginning of t(Y mk1 ) / 2 time window Z s , calculate the pulse signal in the time window Z and the time window Z s The mutual correlation coefficient a between the inner pulse signals Z1 , slide the time window Z backward s , after each sliding, the mutual correlation coefficient a is calculated Zx , we get a set of mutual correlation numbers {a Z1 ,a Z12 ,…,a Zx }, select the time window Z corresponding to the maximum value in the set of correlation coefficients s The time difference △t between the time window Z and the Y sn1 The arrival time is t 12 , where t 12 =t 11 +△t, t 11 and the 12 This is the arrival time point data of the first group; Step S74, select the set Y again mk The second element Y in mk2 With the set Y sn The element Y at the corresponding position sn2 Pairing is performed into the second group and the second group of arrival time point data is calculated, and pairing is performed in sequence until the arrival time point data of the p group is calculated.
8. The method for locating multi-source partial discharge in a high-voltage switch cabinet according to claim 7, characterized in that: The step S8 determines p locations of the same partial discharge source according to the p groups of arrival time point data and the propagation speed V of the pulse signal, including: Step S81, according to the time point t in a set of arrival time point data p1 , the time point t p2 and the propagation speed V of the pulse signal, and calculate the distance between the partial discharge source and the main end sensor measuring point; Step S82, calculating p positioning positions of the same partial discharge source according to p groups of arrival time point data.
Citation Information
Patent Citations
Detection and verification system and method of transformer partial discharge and positioning on-line monitoring assembly
CN102866377A
Partial discharge detection test platform of medium-voltage switch cabinet
CN103176110A
Needle-plate electrode medium space charge and partial-discharge ultrahigh-frequency signal synchronous measurement device
CN103884973A
Multi-band 5G terminal antenna
CN112909511A
Cubical switchboard is ultra -high -frequency antenna for partial discharge monitoring
CN206270450U
