Power transformer winding composite fault realization device and diagnosis method

Abstract

The invention discloses a power transformer winding composite fault realization device and diagnosis method based on a vibration method, and the method comprises the steps: firstly testing a test transformer winding, connecting a transformer winding connection point with a high-frequency high-voltage switch and a high-frequency high-voltage DC power supply, and enabling the high-frequency high-voltage switch to carry out the periodic operation; therefore, the power transformer body structure is excited to vibrate; a high-speed camera is used for shooting vibration of the box wall of the power transformer, and a normalized cross-correlation template matching algorithm is used for extracting vibration signals; secondly, establishing a time-frequency mapping function # imgabs0 # through synchronous extraction change to remove noise interference, performing reconstruction analysis on a signal x0 (t), modulating a vibration signal, establishing a frequency spectrum model X (f, phi), and establishing a frequency spectrum model X (f, phi); obtaining a frequency amplitude sequence and As (f) of the vibration signal and a corresponding frequency phase sequence # imgabs1 #, and calculating energy distribution of different modulation sub-bands by utilizing frequency amplitude sequence self-adaptive frequency band segmentation of a weighted time-frequency variation mode enhancement algorithm; and finally, respectively calculating the transformation rules of the winding under the compound fault operation condition and the normal operation condition through the defined function phase entropy psi (r) and the amplitude-frequency energy function H (f), thereby judging the fault degree of the power transformer winding.

Description

Technical Field

[0001] The present invention is devoted to the technical field of dynamic characteristic analysis and contactless fault detection of power transformers, and specifically relates to a device for realizing composite faults of power transformer windings and a fault degree diagnosis method based on vibration monitoring. Background Art

[0002] Power transformers are key equipment for achieving high-voltage power transmission in power systems, and their operating status is directly related to the reliability and safety of power systems. In actual operation, power transformers are not only subjected to instantaneous load shocks and frequent short-circuit current shocks, but may also be affected by winding deformation and vibration loads under the action of electromagnetic forces. These complex working conditions can cause the winding clamps to loosen or fall off, which in turn causes faults such as axial radial deformation of the winding or damage to the winding insulation layer. The mechanical bearing capacity of the winding decreases accordingly, which may cause partial or overall damage to the winding, and in severe cases may even cause the transformer to stall or be damaged, thereby affecting the stable operation of the UHV power transmission and power supply system. In order to ensure the normal operation of the transformer, the power system needs to perform regular routine inspections on the transformer windings. However, radial deformation of the winding and insulation layer damage are common fault types, and there is currently a lack of diagnostic methods for composite faults (such as radial deformation and insulation layer damage of the winding at the same time), and the existing technology is difficult to comprehensively and accurately identify fault characteristics under different winding states. Therefore, how to achieve dynamic characteristics analysis of complex winding states and accurate diagnosis of fault severity through contactless detection technology is a key technical problem that needs to be solved urgently. Summary of the invention

[0003] The present invention provides a device for realizing compound faults in a power transformer winding. This device realizes radial deformation and surface insulation layer damage at specific positions of the windings of phases A, B, and C through mechanical movement. The electric control box drives the hydraulic drive motor to work, further pushing the first-type hydraulic drive deformation mechanism and the second-type hydraulic drive deformation mechanism to cause different degrees of radial feeding extrusion of the winding deformation ends at the first position, the second position, and the third position, thereby realizing different degrees of radial deformation of the windings. Then, the electric control box controls the movement of the axial movement etching mechanism and the feeding of the etching end of the mechanism to complete the damage of the winding surface insulation layer to different degrees at different positions. The vibration signal on the surface of the power transformer oil tank presents a non-constant curve, which is closely related to the winding state. When the winding is deformed, the insulation layer is damaged, or other faults occur, the waveform and characteristics of the vibration signal will change significantly, and the winding state can be diagnosed by capturing the vibration signal on the surface of the oil tank with a high-speed camera. Therefore, to solve the problems in the prior art, the present invention provides a device for realizing compound faults in a power transformer winding and a method for diagnosing the fault degree based on vibration monitoring, which uses a mechanical device to simulate the radial deformation of the winding and the damage of the surface insulation layer at specific positions, combines with a high-speed camera test system, and further judges the fault degree of the power transformer winding by calculating the phase entropy and the transformation law of the amplitude-frequency energy function on the surface of the power transformer oil tank. The present invention can reliably and effectively realize the fault diagnosis in the compound state of radial deformation of the power transformer winding and damage of the surface insulation layer.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] An apparatus and a diagnostic method for realizing composite faults in a power transformer winding. A mechanical device is used to simulate radial deformation of the winding and damage to the surface insulation layer of the winding under different degrees in the power transformer, and a high-speed camera is combined to realize non-contact acquisition of the vibration signal of the power transformer winding to diagnose the degree of winding faults. It includes the first high-voltage winding, the second low-voltage winding, the third high-voltage winding, the fourth low-voltage winding, the fifth high-voltage winding, the sixth low-voltage winding of the power transformer, the coil pressing parts located on the winding, the bottom support frame of the power transformer winding, the tool placement table, the electric control box, the driving motor of the roller mechanism, the driving sprocket, the chain, the driven sprocket, the left-end support ring mechanism and the right-end support ring mechanism located on the bottom support frame of the power transformer winding, the left-foot support mechanism and the right-foot support mechanism fixed on the left-end support ring mechanism and the right-end support ring mechanism, the upper I-shaped support platform and the lower rectangular support platform fixed on the left-foot support mechanism and the right-foot support mechanism, the roller conveyor mechanism located on the lower rectangular support platform, the roller edge protection frame, the roller, the bearing seat support plate, the auxiliary stable support mechanism, the axial slide rail, the axial slide seat, the axial moving etching mechanism, the end clamping arm, the etching end located on the upper I-shaped support platform, the first type of hydraulic-driven deformation mechanism, the second type of hydraulic-driven deformation mechanism, the hydraulic driving motor, the hydraulic piston cylinder, the deformation end of the winding at the first position, the deformation end of the winding at the second position, the deformation end of the winding at the third position located on the left-foot support mechanism and the right-foot support mechanism, the rotating motor, the V-shaped support seat, the rotating driving gear, the rotating transmission gear, the winding rotating main shaft, the bearing seat, the upper sleeve of the first high-voltage winding, the upper sleeve of the second low-voltage winding, the upper sleeve of the third high-voltage winding, the upper sleeve of the fourth low-voltage winding, the upper sleeve of the fifth high-voltage winding, the upper sleeve of the sixth low-voltage winding, the high-frequency high-voltage DC power supply controlled by a high-frequency high-voltage switch, the high-speed camera, and the signal acquisition device connected to the high-speed camera.

[0006] The present invention also provides a diagnostic method for the apparatus for realizing composite faults in a power transformer winding, including the following steps:

[0007] Step 1: Realize radial deformation of the winding in the power transformer under different degrees through a mechanical device;

[0008] Step 2: Realize damage to the surface insulation layer of the winding in the power transformer under different degrees through a mechanical device;

[0009] Step 3: Conduct machine vision vibration testing and fault diagnosis on the power transformer winding.

[0010] Furthermore, the said Step 1 includes:

[0011] 1) The first high-voltage winding and the second low-voltage winding form the A phase of the power transformer, the third high-voltage winding and the fourth low-voltage winding form the B phase of the power transformer, and the fifth high-voltage winding and the sixth low-voltage winding form the C phase of the power transformer. Place the A phase, B phase, and C phase in the voltage transformer winding composite fault implementation device in sequence, and realize the radial deformation of the winding and the damage of the surface insulation layer of the winding through mechanical movement.

[0012] 2) Place the A phase on the roller conveyor mechanism on the lower rectangular support platform. The electric control box drives the roller mechanism to drive the motor to rotate, realizing the chain drive formed by the driving sprocket, chain, and driven sprocket, and driving the roller conveyor mechanism to realize the axial movement of the A phase. Move it to be coaxially positioned and locked with the winding rotation main shaft to achieve axial fixation.

[0013] 3) The electric control phase controls the hydraulic drive motor to work, driving the first-type hydraulic drive deformation mechanism and the second-type hydraulic drive deformation mechanism located on both sides to move. The hydraulic piston cylinder pushes the winding deformation ends at the first position, the second position, and the third position to radially feed to different degrees, squeezing the A phase to realize different degrees of radial deformation of the winding.

[0014] 4) The hydraulic drive motor rotates in the reverse direction, causing the winding deformation ends at the first position, the second position, and the third position to retract.

[0015] 5) The electric control phase controls the rotating motor located on the bearing seat support plate, driving the gear drive composed of the rotating drive gear and the rotating transmission gear, driving the winding rotation main shaft to rotate, and rotating the A phase to different angles.

[0016] 6) Repeat the above operations to achieve radial deformation at different circumferential positions of the winding.

[0017] Further, the second step includes:

[0018] 1) The electric control box drives the axial movement etching mechanism on the upper I-shaped support platform to move axially along the axial slide rail, so that the axial movement etching mechanism moves to the upper end of the calibration position for damaging the surface insulation layer of the winding.

[0019] 2) Driven by the internal motor of the axial movement etching mechanism, the etching end extends and gradually approaches the surface of the winding insulation layer. The etching end continuously feeds and etches the winding insulation layer at the calibration position, and the end clamping arm prevents the etching end from deflecting during the etching process.

[0020] 3) The electric control box drives the axial movement etching mechanism to move axially along the axial slide rail, and completes the damage of the surface insulation layer of the winding to different degrees at different positions on the same axis by controlling the feed amount of the etching end.

[0021] 4) Electrically control the rotating motor located on the bearing seat support plate, drive the gear transmission composed of the rotating drive gear and the rotating transmission gear, drive the winding rotating main shaft to rotate, and rotate the A phase to different angles.

[0022] 5) Repeat the above steps to achieve the destruction of the surface insulation layer at different positions on the circumferential surface of the winding.

[0023] Further, the third step includes:

[0024] 1) Define the connection point of the upper bushing connecting the first high-voltage winding, the third high-voltage winding, and the fifth high-voltage winding as Y; define the connection point of the upper bushing connecting the second low-voltage winding, the fourth low-voltage winding, and the sixth low-voltage winding as F;

[0025] 2) Connect the transformer winding connection points Y and F to the high-frequency high-voltage switch and the high-frequency high-voltage DC power supply. The high-frequency high-voltage switch performs periodic actions, thereby exciting the vibration of the main body structure of the power transformer.

[0026] 3) Use a high-speed camera to photograph the vibration of the power transformer tank wall, and use the normalized cross-correlation template matching algorithm (NCC) to match and extract the vibration signal frame by frame for the gray-scale image sequence, and transmit the original signal x(t) through the signal acquisition device. The NCC operation formula is as follows, where A is the image of the corresponding matching area of the current image frame, and B is the template area image of the marked point on the power transformer tank wall.

[0027]

[0028] 4) Randomly select a short-time signal x0(t) with at least one complete vibration cycle from the original signal, and establish a time-frequency mapping function by synchronous extraction of changes Remove the interference of noise, perform reconstruction analysis on the signal x0(t), and denote it as The formula is as follows:

[0029] where is the Fourier transform of the window function g, and γ(ω - ω def (t, f)) is the mapping kernel function.

[0030] 5) Modulate the vibration signal to establish a spectrum model X(f, φ), which can be approximately regarded as the result of the convolution superposition of the unit impulse response of the transfer path under the excitation of two types of faults, namely winding deformation and winding insulation layer damage.

[0031]

[0032] where f r is the power grid power frequency of 50HZ, f eis the fundamental vibration frequency of the transformer, 100HZ, f j and f l are the harmonic frequency components contained in the vibration signal, φ k and φ p are the phase angle corresponding to the harmonic frequency, and are the unit impulse responses of the transmission path, A s (f) is the frequency amplitude sequence, is the corresponding frequency phase sequence.

[0033] 6) Use the weighted time-frequency variational mode enhancement algorithm to achieve adaptive frequency band segmentation within the frequency amplitude sequence (A s (f) = 100HZ to 800HZ), calculate the energy distribution p k (f) of different modulation sub-bands d k (f) (k = 1, 2, 3,... K), whose frequency bandwidth is and the center frequency is and represent the lower cut-off frequency domain and the upper cut-off frequency of the k-th sub-band respectively. The calculation formula is as follows: w k (f) = exp(-α k |f k (w) - f k * (w)| 2 )

[0034]

[0035] where w k (f) is the weighting function, and X(f) is the Fourier transform.

[0036] 7) Process the vibration signals under the normal operating conditions of the winding and the combined fault operating conditions of different degrees of deformation and insulation layer damage of the winding according to the above steps. Denote the sub-band energy distribution under the normal operating conditions of the winding as and the frequency phase sequence as

[0037] 8) Construct two groups of n-dimensional vectors from the frequency phase sequences under the combined fault operating conditions and the normal operating conditions of the winding respectively: The distance between the two groups of vectors is expressed as: For a given threshold r, count the number of d[Z ,X i ,X j ≥r and take the ratio to the total number of distances N - 1, denoted as C(r). Take the exponent of C(r), denoted as Ψ(r), and define it as the phase entropy. The calculation formula is as follows:

[0038]

[0039] where α k , β k are phase fluctuation parameters, m is the exponential power, and τ1 is the time variable.

[0040] 9) Define the amplitude-frequency energy function as λ k is the frequency attenuation parameter, P(f) is the cumulative sum of the ratio of the energy distribution p k (f) of each sub-band under the composite fault operation conditions of different degrees of deformation of the winding and damage of the insulating layer to the energy distribution of each sub-band under the normal operation conditions of the winding. Sum.

[0041] 10) Combining the above displacement extraction algorithm, time-frequency variational mode enhancement algorithm, and the defined functions and other processes, finally, by calculating the transformation laws of the phase entropy Ψ(r) and the amplitude-frequency energy function H(f), the fault degree of the power transformer winding is judged as follows:

[0042] If e 4 < Ψ(r) ≤ e 5 , e 3 < H(f) ≤ 6e 3 , it is determined that slight deformation occurs at a few positions of the power transformer winding and the degree of damage to the surface insulating layer is relatively light. At this time, the composite fault degree of the winding does not affect the operation of the power transformer.

[0043] If 2.2e 5 < Ψ(r) ≤ 3.5e 5 , 2.8e 4 < H(f) ≤ 5e 4 , it is determined that slight deformation occurs at multiple positions of the power transformer winding and the surface insulating layer is damaged at multiple positions. At this time, the composite fault degree of the winding will cause increased vibration, and long-term operation will further induce damage to the internal mechanical structure of the power transformer, affecting the service life of the power transformer.

[0044] If 4e 5 ≤ Ψ(r), 5e 4 < H(f), it is determined that serious deformation occurs at multiple positions of the power transformer winding and the degree of damage to the surface insulating layer is relatively heavy. At this time, the composite fault degree of the winding seriously affects the operation of the power transformer, and the winding should be replaced.

[0045] Beneficial effects:

[0046] The present invention uses a mechanical device to simulate the fault in the composite state of the radial deformation of the power transformer winding and the damage of the surface insulation layer, and combines a high-speed camera to realize non-contact acquisition of the transformer vibration signal to achieve winding diagnosis; the present invention can diagnose the composite fault of the radial deformation of the power transformer winding and the damage of the surface insulation layer more reliably and effectively. Description of the Drawings

[0047] Figure 1 It is a schematic diagram of the overall structure of a device for realizing the composite fault of the power transformer winding of the present invention;

[0048] Figure 2 It is a top view of a device for realizing the composite fault of the power transformer winding of the present invention;

[0049] Figure 3 It is a right view of a device for realizing the composite fault of the power transformer winding of the present invention;

[0050] Figure 4 It is a front view of a device for realizing the composite fault of the power transformer winding of the present invention;

[0051] Figure 5 It is a rear view of a device for realizing the composite fault of the power transformer winding of the present invention;

[0052] Figure 6 It is a wiring schematic diagram of the power transformer winding;

[0053] Figure 7 It is a flowchart of the diagnosis method of a device for realizing the composite fault of the power transformer winding of the present invention; Detailed Embodiments

[0054] The following further describes the present invention in detail with reference to the drawings and specific embodiments:

[0055] As Figure 1, as shown in Figures 2, 3, 4, 5, and 6, an apparatus for realizing composite faults in power transformer windings based on vibration method controls the rotation of the main shaft of the transformer winding to drive the winding to rotate around the main shaft. During the rotation, the radial deformation of the winding and the breakage of the surface insulation layer are realized by controlling the feeding of the winding deformation end in the hydraulic drive deformation mechanism and the etching end of the axial movement etching mechanism, including: the first high-voltage winding (1), the second low-voltage winding (2), the third high-voltage winding (3), the fourth low-voltage winding (4), the fifth high-voltage winding (5), the sixth low-voltage winding (6) of the power transformer, the coil pressing member (7) located on the winding, the bottom support frame (8) of the power transformer winding, the tool placement table (9), the electric control box (10), the drum mechanism drive motor (11), the driving sprocket (12), the chain (13), the driven sprocket (14), the left-end support ring mechanism (15) and the right-end support ring mechanism (16) located on the bottom support frame (8) of the power transformer winding, the left-foot support mechanism (17) and the right-foot support mechanism (18) fixed on the left-end support ring mechanism (15) and the right-end support ring mechanism (16), the upper I-shaped support platform (19) and the lower rectangular support platform (20) fixed on the left-foot support mechanism (17) and the right-foot support mechanism (18), the drum conveying mechanism (21) located on the lower rectangular support platform (20), the drum edge protection frame (22), the drum (23), the bearing seat support plate (24), the auxiliary stable support mechanism (25), the axial slide rail (26), the axial slide block (27), the axial movement etching mechanism (28), the end clamping arm (29), the etching end (30) located on the upper I-shaped support platform (19), the first type of hydraulic drive deformation mechanism (31), the second type of hydraulic drive deformation mechanism (32), the hydraulic drive motor (33), the hydraulic piston cylinder (34), the winding deformation end at the first position (35), the winding deformation end at the second position (36), the winding deformation end at the third position (37) located on the left-foot support mechanism (17) and the right-foot support mechanism (18), the rotation motor (38), the V-shaped support seat (39), the rotation drive gear (40), the rotation transmission gear (41), the winding rotation main shaft (42), the bearing seat (43), the upper bushing (44) of the first high-voltage winding (1), the upper bushing (45) of the second low-voltage winding (2), the upper bushing (46) of the third high-voltage winding (3), the upper bushing (47) of the fourth low-voltage winding (4), the upper bushing (48) of the fifth high-voltage winding (5), the upper bushing (49) of the sixth low-voltage winding (6), the high-frequency high-voltage DC power supply (51) controlled by the high-frequency high-voltage switch (50), the high-speed camera (52), and the signal acquisition device (53) connected to the high-speed camera (52).

[0056] Figure 7It is a method for diagnosing composite faults in power transformer windings based on the vibration method. It is characterized by combining a high-speed camera to capture the vibration signal of the transformer tank wall, establishing a time-frequency mapping function through synchronous extraction of changes to remove noise interference, modulating the vibration signal, establishing a spectral model, and finally calculating the transformation rules of the phase entropy and amplitude-frequency energy function, and using this to judge the severity of the composite fault in the winding. The specific steps are as follows:

[0057] 1) Define the connection point of the upper bushings connecting the first high-voltage winding (1), the third high-voltage winding (3), and the fifth high-voltage winding (5) as Y; define the connection point of the upper bushings connecting the second low-voltage winding (2), the fourth low-voltage winding (4), and the sixth low-voltage winding (6) as F;

[0058] 2) Connect the transformer winding connection points Y and F to the high-frequency high-voltage switch (50) and the high-frequency high-voltage DC power supply (51). The high-frequency high-voltage switch (50) performs periodic actions, thereby exciting the vibration of the power transformer body structure.

[0059] 3) Use a high-speed camera to capture the vibration of the power transformer tank wall, and use the normalized cross-correlation template matching algorithm (NCC) to match and extract the vibration signal frame by frame for the gray-scale image sequence, and transmit the original signal x(t) through the signal acquisition device (53). The NCC operation formula is as follows, where A is the image of the corresponding matching area of the current image frame, and B is the template area image of the marked point on the power transformer tank wall.

[0060]

[0061] 4) Randomly select a short-time signal x0(t) with at least one complete vibration cycle from the original signal, and establish a time-frequency mapping function through synchronous extraction of changes to remove the interference of noise, and perform reconstruction analysis on the signal x0(t), denoted as The formula is as follows:

[0062] where is the Fourier transform of the window function g, and γ(ω - ω def (t, f)) is the mapping kernel function.

[0063] 5) Modulate the vibration signal to establish a spectral model X(f, φ), which can be approximately regarded as the result of the convolution superposition of the unit impulse responses of the transfer paths under the excitation of two types of faults, namely winding deformation and winding insulation layer damage, respectively.

[0064]

[0065] where f r is the power grid power frequency of 50HZ, fe is the fundamental vibration frequency of the transformer, 100HZ, f j and f l are the harmonic frequency components contained in the vibration signal, φ k and φ p are the phase angle positions corresponding to the harmonic frequencies. and are the unit impulse responses of the transmission path, A s (f) is the frequency amplitude sequence, is the corresponding frequency phase sequence.

[0066] 6) Use the weighted time-frequency variational mode enhancement algorithm to achieve adaptive frequency band segmentation within the frequency amplitude sequence (A s (f) = 100HZ to 800HZ), and calculate the energy distribution p k (f) of different modulation sub-bands d k (f) (k = 1, 2, 3,... K), whose frequency bandwidth is and the center frequency is and represent the lower cut-off frequency domain and the upper cut-off frequency of the k-th sub-band respectively. The calculation formula is as follows:

[0067]

[0068] where w k (f) is the weighting function, and X(f) is the Fourier transform of.

[0069] 7) Process the vibration signals under the normal operating conditions of the winding, and the combined fault operating conditions of different degrees of deformation and insulation layer damage of the winding respectively according to the above steps. Denote the sub-band energy distribution under the normal operating conditions of the winding as and the frequency phase sequence as

[0070] 8) Construct the frequency phase sequences under the combined fault operating conditions and the normal operating conditions of the winding into two groups of n-dimensional vectors respectively: The distance between the two groups of vectors is expressed as: For a given threshold r, count the number of d[Z i ,X j ≥ r and take the ratio to the total number of distances N - 1, denoted as C(r). Take the exponential of C(r), denoted as Ψ(r), and define it as the phase entropy. The calculation formula is as follows:

[0071]

[0072] where α k , β kis the phase fluctuation parameter, m is the exponential power, and τ1 is the time variable.

[0073] 9) Define the amplitude-frequency energy function as λ k is the frequency attenuation parameter, P(f) is the cumulative sum of the ratio of the energy distribution p k (f) of each sub-band under the compound fault operation condition of different degrees of deformation of the winding and damage of the insulation layer to the energy distribution of each sub-band under the normal operation condition of the winding. Sum.

[0074] 10) By synthesizing the above displacement extraction algorithm, time-frequency variational mode enhancement algorithm, and the defined functions, etc., finally, by calculating the transformation laws of the phase entropy Ψ(r) and the amplitude-frequency energy function H(f), the fault degree of the power transformer winding is judged as follows:

[0075] If e 4 <Ψ(r) ≤ e 5 , e 3 <H(f) ≤ 6e 3 , it is determined that slight deformation occurs at a small number of positions of the power transformer winding and the surface insulation layer is slightly damaged. At this time, the compound fault degree of the winding does not affect the operation of the power transformer.

[0076] If 2.2e 5 <Ψ(r) ≤ 3.5e 5 , 2.8e 4 <H(f) ≤ 5e 4 , it is determined that slight deformation occurs at multiple positions of the power transformer winding and the surface insulation layer is damaged at multiple places. At this time, the compound fault degree of the winding will cause increased vibration, and long-term operation will further induce damage to the internal mechanical structure of the power transformer, affecting the service life of the power transformer.

[0077] If 4e 5 ≤ Ψ(r), 5e 4 <H(f), it is determined that serious deformation occurs at multiple positions of the power transformer winding and the surface insulation layer is severely damaged. At this time, the compound fault degree of the winding seriously affects the operation of the power transformer, and the winding should be replaced.

Claims

1. A device and a method for diagnosing composite faults of power transformer windings, characterized in that: A mechanical device is used to simulate the radial deformation of the winding and the damage of the insulation layer on the winding surface of the power transformer under different degrees, and a high-speed camera is used to realize non-contact collection of the vibration signal of the power transformer winding to realize the diagnosis of the degree of winding fault: including a first high-voltage winding (1) of the power transformer, a second low-voltage winding (2), a third high-voltage winding (3), a fourth low-voltage winding (4), a fifth high-voltage winding (5), a sixth low-voltage winding (6), a coil clamping member (7) located on the winding, a bottom support frame of the power transformer winding (8), a tool placement table (9), an electric control box (10), a roller mechanism driving motor (11), a driving sprocket (12), a chain (13), A movable sprocket (14), a left end supporting wheel ring mechanism (15) and a right end supporting wheel ring mechanism (16) located on a bottom supporting frame (8) of a power transformer winding, a left foot frame mechanism (17) and a right foot frame mechanism (18) fixed on the left end supporting wheel ring mechanism (15) and the right end supporting wheel ring mechanism (16), an upper end profile supporting platform (19) and a lower end rectangular supporting platform (20) fixed on the left foot frame mechanism (17) and the right foot frame mechanism (18), a roller conveying mechanism (21) located on the lower end rectangular supporting platform (20), a roller edge protection frame (22), a roller (23), a bearing seat support plate (24), an auxiliary stabilizing support mechanism ( 25), an axial slide rail (26) located on the upper end of the workpiece support platform (19), an axial slide seat (27), an axial movable etching mechanism (28), an end clamping arm (29), an etching end (30), a first type hydraulic drive deformation mechanism (31) located on the left foot frame mechanism (17) and the right foot frame mechanism (18), a second type hydraulic drive deformation mechanism (32), a hydraulic drive motor (33), a hydraulic piston cylinder (34), a No. 1 winding deformation end (35), a No. 2 winding deformation end (36), a No. 3 winding deformation end (37), a rotating motor (38) located on the bearing seat support plate (24), a V-shaped support seat (3 9), a rotating drive gear (40), a rotating transmission gear (41), a winding rotating main shaft (42), a bearing seat (43), an upper end sleeve (44) of the first high-voltage winding (1), an upper end sleeve (45) of the second low-voltage winding (2), an upper end sleeve (46) of the third high-voltage winding (3), an upper end sleeve (47) of the fourth low-voltage winding (4), an upper end sleeve (48) of the fifth high-voltage winding (5), an upper end sleeve (49) of the sixth low-voltage winding (6), a high-frequency high-voltage direct current power supply (51) controlled by a high-frequency high-voltage switch (50), a high-speed camera (52), and a signal acquisition device (53) connected to the high-speed camera (52).

2. A device and method for realizing composite fault of power transformer winding according to claim 1, characterized in that: The following steps are involved: Step 1: Achieving radial deformation of the winding of the power transformer to different degrees through a mechanical device; Step 2: Using a mechanical device to achieve different degrees of damage to the surface insulation layer of the winding of the power transformer; Step 3: Conduct machine vision vibration testing and fault diagnosis of power transformer windings.

3. The diagnostic method according to claim 2, characterized in that The step one comprises: 1) The first high-voltage winding (1) and the second low-voltage winding (2) constitute the A phase of the power transformer, the third high-voltage winding (3) and the fourth low-voltage winding (4) constitute the B phase of the power transformer, and the fifth high-voltage winding (5) and the sixth low-voltage winding (6) constitute the C phase of the power transformer. The A phase, the B phase, and the C phase are sequentially placed in a voltage transformer winding composite fault realization device, and the radial deformation of the winding and the destruction of the insulation layer on the winding surface are realized through the movement of the mechanism. 2) Phase A is placed on a roller conveying mechanism (21) on a rectangular support platform (20) at the lower end, and an electric control box (10) drives a roller mechanism driving motor (11) to rotate, thereby realizing a chain drive formed by a driving sprocket (12), a chain (13) and a driven sprocket (14), and driving the roller conveying mechanism (21) to realize axial movement of phase A, and move phase A to be coaxially positioned and locked with a winding rotating main shaft (42), thereby realizing axial fixation. 3) The electric control phase (10) controls the hydraulic drive motor (33) to work, driving the first type hydraulic drive deformation mechanism (31) and the second type hydraulic drive deformation mechanism (32) located on both sides to move, and the hydraulic piston cylinder (34) pushes the first winding deformation end (35), the second winding deformation end (36), and the third winding deformation end (37) to radially feed to different degrees, squeeze the A phase, and realize radial deformation of the winding to different degrees. 4) The hydraulic drive motor (33) runs in the reverse direction, causing the No. 1 winding deformation end (35), the No. 2 winding deformation end (36), and the No. 3 winding deformation end (37) to retract. 5) The electric control phase (10) controls the rotating motor (38) located on the bearing seat support plate (24), drives the rotating drive gear (40) and the rotating transmission gear (41) to form a gear transmission, drives the winding rotating main shaft (42) to rotate, and rotates the A phase to different angles. 6) Repeat the above operation to achieve radial deformation at different circumferential positions of the winding.

4. The diagnostic method according to claim 3, characterized in that: The second step comprises: 1) The electric control box (10) drives the axially movable etching mechanism (28) located on the upper end of the industrial support platform (19) to move axially along the axial slide rail (26), so that the axially movable etching mechanism (28) moves to the upper end of the marked position where the insulation layer on the surface of the winding is damaged. 2) The axially movable etching mechanism (28) is driven by an internal motor so that the etching tip (30) extends and gradually approaches the surface of the winding insulation layer. The etching tip (30) continuously feeds and etches the winding insulation layer at the marked position. The tip clamping arm (29) prevents the etching tip (30) from being deflected during the etching process. 3) The electric control box (10) drives the axially movable etching mechanism (28) to move axially along the axial slide rail (26), and controls the feed rate of the etching end (30) to achieve different degrees of destruction of the insulation layer on the winding surface at different positions on the same axis. 4) The electric control phase (10) controls the rotating motor (38) located on the bearing seat support plate (24), drives the rotating drive gear (40) and the rotating transmission gear (41) to form a gear transmission, drives the winding rotating main shaft (42) to rotate, and rotates the A phase to different angles. 5) Repeat the above steps to destroy the surface insulation layer at different positions on the circumferential surface of the winding.

5. The diagnostic method according to claim 4, characterized in that: The step three comprises: 1) defining a connection point of the upper end bushings connecting the first high-voltage winding (1), the third high-voltage winding (3), and the fifth high-voltage winding (5) as Y; defining a connection point of the upper end bushings connecting the second low-voltage winding (2), the fourth low-voltage winding (4), and the sixth low-voltage winding (6) as F; 2) The transformer winding connection point Y and the connection point F are connected to a high-frequency high-voltage switch (50) and a high-frequency high-voltage direct current power source (51), and the high-frequency high-voltage switch (50) performs periodic actions, thereby exciting the vibration of the power transformer body structure. 3) A high-speed camera is used to shoot the vibration of the power transformer box wall, and a normalized cross-correlation template matching algorithm (NCC) is used to match the grayscale image sequence frame by frame to extract the vibration signal, and the original signal x(t) is transmitted through the signal acquisition device (53). The NCC operation formula is as follows, where A is the matching area image corresponding to the current image frame, and B is the template area image of the selected power transformer box wall calibration point. 4) Randomly select a short-time signal x0(t) with at least one complete vibration cycle from the original signal, and establish a time-frequency mapping function by synchronously extracting changes Remove the interference of noise and reconstruct the signal x0(t), which is recorded as The formula is as follows: in is the Fourier transform of the window function g, γ(ω-ω def (t,f)) is the mapping kernel function. 5) Yes The vibration signal is modulated to establish the spectrum model X(f,φ), which can be approximately regarded as the result of the convolution superposition of the unit impulse response of the transmission path under the two fault excitations of winding deformation and winding insulation layer damage. where f r The power frequency of the power grid is 50HZ, f e The transformer vibration fundamental frequency is 100HZ, f j and f l is the harmonic frequency component contained in the vibration signal, φ k and φ p is the phase angle corresponding to the harmonic frequency, and is the unit impulse response of the transmission path, A s (f) is the frequency amplitude sequence, is the corresponding frequency phase sequence. 6) Using the weighted time-frequency variational modal enhancement algorithm to realize the frequency amplitude sequence (A s (f) = 100 Hz ~ 800 Hz) within the adaptive frequency band segmentation, calculate the different modulation sub-band d k (f) Energy distribution p k (f)(k=1,2,3,…K), its frequency bandwidth is The center frequency is and Respectively represent the lower cutoff frequency domain and upper cutoff frequency of the kth sub-band. The calculation formula is as follows: w k (f)=exp(-α k |f k (w)-f k * (w)| 2 ) where w k (f) is the weighting function, X(f) is The Fourier transform of . 7) According to the above steps, the vibration signals under normal operation of the winding and under the composite fault operation of the winding with different degrees of deformation and insulation layer damage are processed respectively. The sub-band energy distribution under normal operation of the winding is denoted as The frequency phase sequence is 8) The frequency phase sequence under the winding composite fault operating condition and the normal operating condition Construct two sets of n-dimensional vectors respectively: The distance between two sets of vectors is expressed as: For a given threshold r, the statistic d[Z i ,X j ]≥r and the ratio of the total number of distances N-1 is recorded as C(r), and C(r) is exponentially calculated as Ψ(r), which is defined as phase entropy. The calculation formula is as follows: where α k , β k is the phase fluctuation parameter, m is the exponential power, and τ1 is the time variable. 9) Define the amplitude-frequency energy function as λ k is the frequency attenuation parameter, P(f) is the energy distribution of each sub-band under the condition of winding deformation and insulation layer damage composite fault operation. k (f) and the energy distribution of each sub-band under normal winding operation conditions are: The cumulative sum of ratios. 10) Combining the above displacement extraction algorithm, time-frequency variational mode enhancement algorithm and the defined function, the fault degree of the power transformer winding is finally determined by calculating the transformation law of the phase entropy Ψ(r) and the amplitude-frequency energy function H(f), as shown below: 4 <Ψ(r)≤e 5 , e 3 <H(f)≤6e 3 It is determined that slight deformation occurs in a few locations of the power transformer winding and the degree of damage to the surface insulation layer is relatively light. At this time, the degree of the winding composite fault does not affect the operation of the power transformer. If 2.2e 5 <Ψ(r)≤3.5e 5 , 2.8e 4 <H(f)≤5e 4 It is determined that the power transformer windings have slight deformations in many places and the insulation layer on the surface is damaged in many places. At this time, the degree of complex winding failure will cause the vibration to intensify. Long-term operation will further induce damage to the internal mechanical structure of the power transformer and affect the service life of the power transformer. If 4e 5 ≤Ψ(r),5e 4 <H(f), it is determined that the power transformer winding has serious deformation in many places and the surface insulation layer is severely damaged. At this time, the degree of complex fault of the winding seriously affects the operation of the power transformer, and the winding should be replaced.

Images