Diagnostic positioning method for discharge in transformer oil based on electromagnetic-pressure dual-mode fusion

Abstract

The invention belongs to the technical field of power equipment fault diagnosis, and discloses an electromagnetism-pressure bimodal fusion transformer in-oil discharge diagnosis and positioning method, which comprises the following steps of: acquiring an electromagnetic signal and a pressure wave signal in real time according to more than three electromagnetism-pressure integrated sensing components mounted at different positions in a valve to be maintained; determining the time when the electromagnetic wave and the impact pressure wave respectively reach each electromagnetic-pressure integrated sensing assembly, and further determining the distance from the discharge point to each electromagnetic-pressure integrated sensing assembly; constructing a positioning equation set according to the distance from the discharge point to each electromagnetic-pressure integrated sensing assembly; and solving the positioning equation set, and determining the position of the discharge point. The invention provides a method for jointly positioning a discharge point based on an electromagnetic wave signal and an impact pressure signal for the first time; and the distance relationship between the discharge point and each electromagnetic-pressure integrated sensing assembly is established by using the time difference between the discharge point and each electromagnetic-pressure integrated sensing assembly, so that a positioning equation set is constructed, and accurate positioning of the discharge point can be realized.

Description

Technical Field

[0001] This invention belongs to the field of power equipment fault diagnosis technology, and relates to transformer oil discharge diagnosis, and particularly to a method for diagnosing and locating transformer oil discharge by fusing electromagnetic and pressure modes. Background Technology

[0002] As a core piece of equipment in the power system, the transformer's internal insulating oil not only serves for cooling but also plays a crucial role in insulation. However, discharge in the transformer oil is an important indicator of early equipment failure, typically caused by aging, contamination, cracks in the insulation material, or insulator failure. Timely detection of discharge in the oil is essential for identifying potential insulation risks, preventing equipment damage, and ensuring the safe and stable operation of the power system.

[0003] During the discharge process of electrical equipment, various physical phenomena occur, such as charge exchange, electromagnetic radiation, and sound wave generation. Existing methods for detecting discharge in oil mainly include pulse current methods, radio interference methods, ultra-high frequency methods, and gas chromatography analysis. While the pulse current method can quantitatively detect the discharge quantity, its anti-interference capability is weak and it is easily affected by electromagnetic noise. The radio interference method can only qualitatively detect the discharge and cannot determine the discharge intensity. Although the ultra-high frequency method can avoid some electrical interference, it also cannot determine the discharge intensity. Gas chromatography, due to its long oil-gas separation time, cannot diagnose discharge in oil in a timely manner. All these methods have limitations and cannot achieve accurate online detection of discharge in oil.

[0004] Furthermore, in actual operation, strong electromagnetic interference exists inside the transformer, which greatly reduces the accuracy and sensitivity of discharge detection and easily leads to misjudgments. Therefore, how to accurately extract discharge signals under strong interference environments and provide reliable information for insulation diagnosis is a key issue in online transformer discharge monitoring. Summary of the Invention

[0005] To address the shortcomings of existing methods for detecting discharge in transformer oil, such as weak anti-interference capabilities and inability to accurately locate discharge defects, this invention innovatively proposes a joint diagnosis and precise location method for discharge in transformer oil based on electromagnetic-pressure dual-mode signal fusion. By simultaneously capturing the electromagnetic radiation signal and the hydraulic-electric impact pressure wave signal generated by the discharge in transformer oil, a dual-signal characteristic parameter correlation model is established to achieve joint diagnosis and three-dimensional precise location of discharge events. This provides a novel monitoring technology with both high sensitivity and high reliability for transformer insulation condition assessment.

[0006] To achieve the above objectives, the present invention adopts the following technical solutions.

[0007] This invention provides a method for diagnosing and locating discharge in transformer oil using a fusion of electromagnetic and pressure modes, comprising the following steps:

[0008] S1, based on three or more electromagnetic-pressure integrated sensing components installed in different positions inside the valve to be inspected, collects electromagnetic signals and pressure wave signals in real time.

[0009] S2, determine the time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component and the time when the shock pressure wave arrives at each electromagnetic-pressure integrated sensing component, and then determine the distance from the discharge point to each electromagnetic-pressure integrated sensing component.

[0010] S3. Based on the distance from the discharge point to each electromagnetic-pressure integrated sensing component, a set of positioning equations is constructed.

[0011] S4. Solve the positioning equations to determine the location of the discharge point.

[0012] In step S1 above, the electromagnetic-pressure integrated sensing component includes a housing and a high-frequency current sensor and a piezoelectric high-frequency pressure sensor stacked together and installed inside the housing.

[0013] In step S2 above, the distance d from the discharge point to the i-th integrated electromagnetic-pressure sensing component is calculated according to the following formula. i :

[0014] ;

[0015] Among them, t ei t represents the time it takes for the electromagnetic wave to reach the i-th electromagnetic-pressure integrated sensing component; pi v is the time it takes for the shock pressure wave to reach the i-th electromagnetic-pressure integrated sensing component; e The speed at which electromagnetic waves propagate in transformer oil is close to the speed of light; v p This refers to the propagation speed of the shock pressure wave in the transformer oil.

[0016] v p It will change with oil temperature, and the specific result can be determined using the following formula:

[0017] ;

[0018] Where α is the temperature correction factor; T is the oil temperature; This is a reference temperature.

[0019] Step S2 above includes the following sub-steps:

[0020] S21, Based on the set electromagnetic wave trigger threshold, determine the timing trigger start point of all electromagnetic-pressure integrated sensing components;

[0021] S22, starting from the self-timer trigger point, within the adjustable monitoring window, determines the time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component based on the electromagnetic wave adaptive threshold.

[0022] S23, starting from the self-timer trigger point, within the adjustable monitoring window, the time for the shock pressure wave to reach each electromagnetic-pressure integrated sensing component is determined based on the shock pressure wave adaptive threshold.

[0023] S24. Based on the time it takes for the electromagnetic wave to reach each electromagnetic-pressure integrated sensing component and the time it takes for the impact pressure wave to reach each electromagnetic-pressure integrated sensing component, the distance from the discharge point to each electromagnetic-pressure integrated sensing component is calculated.

[0024] In step S22 above, the arrival time of electromagnetic waves at each electromagnetic-pressure integrated sensing component is determined according to the following steps:

[0025] (A1) Calculate the time period T before the adjustable monitoring window or the start of the timing trigger. pre Internal electromagnetic signal V e,i The root mean square noise level N of (t) rms,e,i ;

[0026] (A2) Based on electromagnetic signal V e,i The root mean square noise level N of (t) rms,e,i Construct an adaptive threshold V′ for electromagnetic waves th,e ;

[0027] (A3) The time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component is the time when the corresponding signal amplitude first exceeds the adaptive threshold V′. th,e The moment t ei .

[0028] To improve time accuracy, when determining t ei The area is refined using linear interpolation or cubic spline interpolation, with the time corresponding to the highest amplitude point being taken as the time when the electromagnetic wave arrives at the corresponding electromagnetic-pressure integrated sensing component.

[0029] In step S23 above, the time it takes for the shock pressure wave to reach each electromagnetic-pressure integrated sensing component is determined according to the following steps:

[0030] (B1) Calculate the time period T before the adjustable monitoring window or the start of the timing trigger. pre Internal shock pressure wave signal V p,i The root mean square noise level N of (t) rms,p,i ;

[0031] (B2) Based on the impact pressure wave signal V p,i The root mean square noise level N of (t) rms,p,iConstruct an adaptive threshold V for shock pressure waves th,p ;

[0032] (B3) The time when the shock pressure wave arrives at each electromagnetic-pressure integrated sensing component is the time when the corresponding signal amplitude first exceeds the shock pressure wave adaptive threshold V. th,p The moment t pi .

[0033] To improve time accuracy, when determining t pi The area is refined using linear interpolation or cubic spline interpolation, with the time corresponding to the highest amplitude point being taken as the time when the shock pressure wave arrives at the corresponding electromagnetic-pressure integrated sensing component.

[0034] In step S3 above, for n integrated electromagnetic-pressure sensing components, a set of n distance equations between the discharge point and the n integrated electromagnetic-pressure sensing components is constructed; then, by using the finite difference method, the quadratic terms in each equation are eliminated to obtain a set of n-1 linear equations.

[0035] The distance between the discharge point and the i-th electromagnetic-pressure integrated sensing component is defined as:

[0036] ;

[0037] Where (x,y,z) represents the location of the discharge point, (x i ,y i ,z i ) represents the position of the i-th electromagnetic-pressure integrated sensing component, and d i Let be the distance between the discharge point and the i-th electromagnetic-pressure integrated sensing component.

[0038] The i-th integrated electromagnetic-pressure sensing component and the j-th integrated electromagnetic-pressure sensing component, after eliminating quadratic terms using the difference method, are expressed as follows:

[0039] .

[0040] In step S4 above, the location of the discharge point is obtained by substituting, eliminating, or using a matrix method. This method allows some variables to be expressed as a function of other variables.

[0041] The rationality of the solution can also be verified by geometric constraints, eliminating invalid solutions that are beyond the actual physical range (such as the actual size of the valve to be repaired).

[0042] Through the above calculation process, the three-dimensional spatial coordinates (x, y, z) of the discharge source can be accurately obtained, achieving high-precision fault location.

[0043] Therefore, this invention employs a built-in integrated electromagnetic-pressure sensor design. By integrating a high-frequency current sensor and a piezoelectric high-frequency pressure sensor into the same measurement unit, and utilizing the natural shielding effect of the transformer's metal casing, accurate acquisition of pure electromagnetic signals is achieved. In terms of signal processing, this invention innovatively establishes a multi-dimensional feature correlation model between electromagnetic pulse signals and hydraulic-electric impact pressure waves. By analyzing the differences and correlations in the time-domain characteristics of the two signals, environmental noise interference is effectively reduced, and the signal-to-noise ratio is significantly improved. Regarding the positioning algorithm, it fully leverages the advantages of the fast propagation speed of electromagnetic signals and the slow attenuation of pressure wave signals. Using a time-difference positioning method combined with a three-dimensional spatial calculation algorithm, precise positioning of the discharge source can be achieved with only three measurement points. Compared to traditional single-signal detection methods, this invention not only solves the problems of a large number of sensors and complex installation but also achieves a dual improvement in detection sensitivity and positioning accuracy, providing a novel technical solution for online monitoring and fault early warning of discharge in transformer oil.

[0044] Compared with existing technologies, the electromagnetic-pressure dual-mode fusion method for diagnosing and locating discharge in transformer oil provided by this invention has the following advantages:

[0045] 1) This invention proposes for the first time a method for locating discharge points based on the joint use of electromagnetic wave signals and impact pressure signals; by utilizing the time difference between the two, the distance relationship between the discharge point and each electromagnetic-pressure integrated sensing component is established, and then a set of positioning equations is constructed, which can achieve accurate positioning of the discharge point;

[0046] 2) This invention integrates a high-frequency current sensor and a pressure sensor into the same measurement unit, which solves the problem of a large number of sensors and complex installation in traditional single signal detection methods. At the same time, it utilizes the natural shielding effect of the transformer's metal shell to achieve accurate acquisition of pure electromagnetic signals.

[0047] 3) This invention establishes a multidimensional feature correlation model between electromagnetic pulse signals and electrohydraulic shock pressure waves. By combining multimodal signals, environmental noise interference is effectively reduced and the signal-to-noise ratio is significantly improved. Furthermore, under oil flow conditions, the separation characteristics of the shock pressure wave characteristic frequency band (10-100 kHz) and the oil flow noise frequency band (<5 kHz) are utilized to achieve effective signal extraction.

[0048] 4) This invention employs a multimodal spatiotemporal intersection positioning algorithm, which requires only 3 measurement points to achieve accurate positioning of the discharge source. Compared with the traditional single signal positioning method, it achieves a dual improvement in detection sensitivity and positioning accuracy.

[0049] 5) This invention provides a novel technical solution for online monitoring and fault early warning of discharge in transformer oil. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating the electromagnetic-pressure dual-mode fusion method for diagnosing and locating discharge in transformer oil provided in an embodiment of the present invention.

[0051] Figure 2 This is a schematic diagram of an integrated electromagnetic-pressure sensing component; where 1-shell; 2-pressure wave receiver; 3-external thread;

[0052] Figure 3 A flowchart illustrating the steps for determining the distance from the discharge point to each integrated electromagnetic-pressure sensing component;

[0053] Figure 4 The electromagnetic wave and shock pressure wave signals collected by the first sensing component in the experimental example of this invention;

[0054] Figure 5 The electromagnetic wave and impact pressure wave signals collected by the second sensing component in the experimental example of this invention;

[0055] Figure 6 These are the electromagnetic wave and impact pressure wave signals collected by the third sensing component in the experimental example of this invention. Detailed Implementation

[0057] The technical solutions of various embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] This invention employs a multimodal spatiotemporal intersection localization (M-STIL) algorithm to achieve three-dimensional spatial positioning of the discharge source. Utilizing the physical difference between the high propagation speed of electromagnetic signals (approaching the speed of light) and the slow propagation speed but low attenuation coefficient of pressure wave signals, a high-precision three-dimensional spatial positioning can be achieved using a combined electromagnetic-pressure wave ranging method with only three measurement points. Compared to traditional single-modal positioning methods, such as purely acoustic positioning methods based on time difference of arrival (TDOA), or the electro-acoustic combined positioning method commonly used in engineering applications combining external current transformers and pressure sensors, this invention significantly reduces the false triggering rate under strong electromagnetic interference environments, effectively improving the reliability and measurement accuracy of the positioning system. The following detailed explanation of the invention's method is provided with reference to specific embodiments. Example

[0059] This embodiment provides a method for diagnosing and locating discharge in transformer oil using a fusion of electromagnetic and pressure modes, such as... Figure 1 As shown, it includes the following steps:

[0060] S1, based on three or more electromagnetic-pressure integrated sensing components installed in different positions inside the valve to be inspected, collects electromagnetic signals and pressure wave signals in real time.

[0061] In this embodiment, the integrated electromagnetic-pressure sensing component used is, for example... Figure 2 As shown, it includes a housing 1, and a high-frequency current sensor and a piezoelectric high-frequency pressure sensor stacked together inside the housing. The pressure wave receiving end 2 of the piezoelectric high-frequency pressure sensor extends outside the housing 1.

[0062] The housing of the integrated electromagnetic-pressure sensing component is made of stainless steel with high magnetic permeability (such as ferritic stainless steel, martensitic stainless steel, etc.). This not only provides structural support, but more importantly, it utilizes the natural shielding effect of the transformer's metal housing to effectively suppress strong external electromagnetic interference, thereby ensuring that the sensor can only accurately acquire pure electromagnetic signals from inside the transformer.

[0063] The high-frequency current sensor detects transient electromagnetic pulse signals generated by discharges in transformer oil through a built-in microstrip antenna structure. To enhance signal reception, the microstrip antenna is manufactured using multilayer printed circuit board technology and has wide bandwidth characteristics (covering a frequency band of 100 kHz-1 GHz), enabling it to capture electromagnetic signals generated by different types of discharges.

[0064] The piezoelectric high-frequency pressure sensor is based on piezoelectric ceramic material and has extremely high sensitivity and fast response characteristics (response time <1 μs) to the liquid electro-impact pressure wave signal generated by the instantaneous discharge in transformer oil. It can accurately convert weak pressure wave signals into charge signals, and directly convert them into standard voltage signals (0-5V) by the built-in charge amplifier, which facilitates subsequent signal processing and analysis.

[0065] The housing 1 of the integrated electromagnetic-pressure sensing assembly has an external thread 3 at the end near the piezoelectric high-frequency pressure sensor, which is then threaded onto a flange. The integrated electromagnetic-pressure sensing assembly is embedded into the inner wall of the transformer maintenance valve via the flange, enabling oil-free venting and non-destructive installation. The end of the housing 1 furthest from the piezoelectric high-frequency pressure sensor extends out of the transformer maintenance valve and transmits the acquired data to a host computer via a connecting lead. The size and shape of the integrated electromagnetic-pressure sensing assembly are customized according to the internal space of the transformer to ensure convenient installation without affecting the normal operation of the transformer. The housing of the integrated electromagnetic-pressure sensing assembly is made of high-strength insulating material to ensure safety and reliability in the high-voltage operating environment of the transformer.

[0066] S2, determine the time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component and the time when the shock pressure wave arrives at each electromagnetic-pressure integrated sensing component, and then determine the distance from the discharge point to each electromagnetic-pressure integrated sensing component.

[0067] For each i-th integrated electromagnetic-pressure sensing component that successfully captures the dual-modal signal, this step calculates the time difference Δt between the arrival of the electromagnetic signal and the pressure wave. i To determine the distance d from the discharge point to the sensor. i .

[0068] like Figure 3 As shown, step S2 includes the following sub-steps:

[0069] S21, based on the set electromagnetic wave trigger threshold, determine the timing trigger start point of all electromagnetic-pressure integrated sensing components.

[0070] When any electromagnetic signal V e,i (t) Instantaneous amplitude | V e,i (t)| First time exceeding the preset electromagnetic wave trigger threshold V th,e Immediately trigger synchronized timing across all channels, recorded as t=0 s, and open an adjustable monitoring window W of 200 μs. mon .

[0071] S22, starting from the self-timer trigger point, within the adjustable monitoring window, determines the time when electromagnetic waves arrive at each electromagnetic-pressure integrated sensing component based on the electromagnetic wave adaptive threshold.

[0072] Here, the arrival time of electromagnetic waves at each electromagnetic-pressure integrated sensing component is determined according to the following steps:

[0073] (A1) Calculate the time period T before the timing trigger start point. pre Internal electromagnetic signal V e,i The root mean square noise level N of (t) rms,e,i ,Right now:

[0074] (1);

[0075] Among them, t start t trig They are time periods T pre The start and end times.

[0076] (A2) Based on electromagnetic signal V e,i The root mean square noise level N of (t) rms,e,i Construct an adaptive threshold V′ for electromagnetic waves th,e ,Right now:

[0077] V′ th,e = K e ×N rms,e,i (2);

[0078] Among them, K e Let K be an empirical coefficient. e =5.

[0079] (A3) The time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component is the time when the corresponding signal amplitude first exceeds the adaptive threshold V′. th,e The moment t ei ,Right now:

[0080] (3).

[0081] To improve time accuracy, when determining t ei The area is refined using linear interpolation or cubic spline interpolation, with the time corresponding to the highest amplitude point being taken as the time when the electromagnetic wave arrives at the corresponding electromagnetic-pressure integrated sensing component.

[0082] S23, starting from the self-timer trigger point, within the adjustable monitoring window, determines the time when the shock pressure wave arrives at each electromagnetic-pressure integrated sensing component based on the shock pressure wave adaptive threshold.

[0083] Here, the arrival time of the shock pressure wave at each electromagnetic-pressure integrated sensing component is determined according to the following steps:

[0084] (B1) Calculate the time period T′ before the timing trigger start point. pre Internal shock pressure wave signal V p,i The root mean square noise level N of (t) rms,p,i ,Right now:

[0085] (4);

[0086] Where, t′ start , t′ trig They are time periods T′ pre The start and end times.

[0087] (B2) Based on the impact pressure wave signal V p,i The root mean square noise level N of (t) rms,p,i Construct an adaptive threshold V for shock pressure waves th,p ,Right now:

[0088] V th,p = K p ×N rms,p,i (5);

[0089] Among them, K pLet K be an empirical coefficient. p =5.

[0090] (B3) The time when the shock pressure wave arrives at each electromagnetic-pressure integrated sensing component is the time when the corresponding signal amplitude first exceeds the shock pressure wave adaptive threshold V. th,p The moment t pi ,Right now:

[0091] (6).

[0092] To improve time accuracy, when determining t pi The area is refined using linear interpolation or cubic spline interpolation, with the time corresponding to the highest amplitude point being taken as the time when the shock pressure wave arrives at the corresponding electromagnetic-pressure integrated sensing component.

[0093] S24. Based on the time it takes for the electromagnetic wave to reach each electromagnetic-pressure integrated sensing component and the time it takes for the impact pressure wave to reach each electromagnetic-pressure integrated sensing component, the distance from the discharge point to each electromagnetic-pressure integrated sensing component is calculated.

[0094] The distance d from the discharge point to the i-th electromagnetic-pressure integrated sensing component is calculated using the following formula. i :

[0095] (7).

[0096] If the calculated distance d i If the data point exceeds the maximum geometric dimensions of the transformer cavity, it is considered invalid.

[0097] S3. Based on the distance from the discharge point to each electromagnetic-pressure integrated sensing component, a set of positioning equations is constructed.

[0098] Distance information d obtained based on single-point ranging i The localization problem can be transformed into finding the intersection point of multiple spheres in three-dimensional space. The spatial coordinates of the power source under test are defined as (x, y, z), and the coordinates of the i-th sensor are (x, y, z). i ,y i ,z i The Euclidean distance d between the discharge point and the i-th sensor. i The following geometric relationship must be satisfied:

[0099] (8);

[0100] Where (x,y,z) represents the location of the discharge point, (x i ,y i ,z i ) represents the position of the i-th electromagnetic-pressure integrated sensing component, and di Let be the distance between the discharge point and the i-th electromagnetic-pressure integrated sensing component.

[0101] When there are n ≥ 3 effective sensing component nodes in the system, a system of equations consisting of n nonlinear equations can be formed.

[0102] Expanding the above equation algebraically, we get:

[0103] (9).

[0104] After sorting, it can be represented as:

[0105] (10).

[0106] The finite difference method can effectively eliminate quadratic terms in each equation. Specifically, it uses the first equation as a baseline and subtracts it from the remaining equations:

[0107] (11).

[0108] After the difference operation, all quadratic terms are canceled out, resulting in a linear equation for (x, y, z):

[0109] (12).

[0110] For a system of equations containing n equations, this method can yield n-1 linear equations, thus transforming the original nonlinear system of equations into a linear system of equations.

[0111] S4. Solve the positioning equations to determine the location of the discharge point.

[0112] The obtained system of linear equations can be solved using various methods, such as expressing some variables as functional relationships between other variables; for example:

[0113] (1) Substitution method: Solve one equation to obtain a function expression of one variable as other variables, and then substitute it into another equation to solve step by step;

[0114] (2) Elimination method: Eliminate a variable by appropriate linear combination to reduce the dimension of the system of equations;

[0115] (3) Matrix method: The system of equations is expressed in matrix form and solved by Gaussian elimination or matrix inversion.

[0116] For three valid sensing component nodes, taking the substitution method as an example, substituting the obtained variable relationship expression back into any of the original spherical equations yields an equation containing only a single variable. For example, substituting the expressions for y(x) and z(x) into the first spherical equation:

[0117] (13).

[0118] Expanding this will result in a quadratic equation in x, in its standard form:

[0119] (14);

[0120] Use the quadratic formula Solve according to the discriminant. Different situations:

[0121] when When the equation has two real solutions, corresponding to the two intersection points of the two spheres;

[0122] when When the equation has a unique real solution, it indicates that the two spheres are tangent;

[0123] when When the equation has no real solutions, it indicates that the spheres have no intersection points.

[0124] The rationality of the solution can also be verified through geometric constraints, eliminating invalid solutions that exceed the actual physical range (such as the actual size of the valve to be inspected). For positioning methods involving multiple sensing components, the residuals and confidence levels of the solutions can be further evaluated to ensure the reliability of the positioning results.

[0125] Experimental Example

[0126] This experiment was conducted in a standard transformer oil tank. First, a needle plate electrode was set at the origin O(0,0,0) as a standard discharge source, and a right-hand rectangular coordinate system (x,y,z) was established as the basis for subsequent positioning.

[0127] Before the experiment begins, basic parameters such as oil temperature and insulation strength are measured and recorded.

[0128] Measure the oil temperature T using a standard thermometer; measure the breakdown voltage U using an insulating oil dielectric strength tester. b Record the ambient temperature, oil temperature T, and insulation strength U. b To ensure compliance with testing standards.

[0129] Determining the velocity of sound v in transformer oil based on oil temperature T. s and electromagnetic wave velocity v e And at the speed of sound v s As the wave velocity of the impact pressure wave v p .

[0130] (15);

[0131] Where α is the temperature correction factor, usually taken as α≈2.5 m / (s·℃), and T0 is the reference temperature T0=25℃.

[0132] (16);

[0133] Where c is the speed of light. is the relative permittivity of the transformer oil.

[0134] Next, the integrated electromagnetic-pressure sensing components were installed. To ensure measurement accuracy, three integrated electromagnetic-pressure sensing components were arranged at different positions from the origin within the oil tank, with their installation coordinates set as follows: P1(0.15, 0, 0.1), P2(0.09, 0.12, 0.4), and P3(-0.12, 0.09, -0.5).

[0135] The integrated electromagnetic-pressure sensing unit employs an oil-resistant, sealed structure and is fixed to the inner wall of the oil tank via a flange. Shielded twisted-pair cables are used for the connection leads to reduce interference. Real-time data acquisition is performed using three integrated electromagnetic-pressure sensing units, and the acquisition results are as follows: Figures 4-6 As shown.

[0136] Let the spatial coordinates of the discharge point P0 be (x, y, z), and its Euclidean distances to each sensor satisfy the set relation:

[0137] (17).

[0138] Based on the electromagnetic wave and impact pressure wave signals collected by the integrated electromagnetic-pressure sensing components, the distance between each integrated electromagnetic-pressure sensing component and the discharge point is determined according to the previously given step S2.

[0139] The propagation speed v of electromagnetic wave signals in transformer oil e ≈2×10 8 m / s, the average time to reach the sensing component Pᵢ is t ei ≈0.

[0140] Following step S23 above, the measured value of the shock pressure wave propagation time is t. p1 =1.086×10 -4 s, t p2 =1.272×10 -4 s, t p3 =1.797×10 -4 s corresponds to calculated distances d1=0.1520 m, d2=0.1781 m, and d3=0.2516 m.

[0141] Based on measured data, a set of positioning equations is established:

[0142] (18);

[0143] Expanding equation (18), we get:

[0144] (19);

[0145] Using P1 as the reference node, nonlinear terms are eliminated through finite difference. ,get:

[0146] (20);

[0147] (twenty one);

[0148] Equation (20) can be rearranged as follows:

[0149] (twenty two);

[0150] Substituting the expression in equation (22) into equation (21), we get:

[0151] (twenty three);

[0152] Right now:

[0153] (twenty four);

[0154] Let x represent y and z. Substitute the expression for y into the expression for z:

[0155] (25);

[0156] Substituting equations (24) and (25) into equation (19-(a)), we get:

[0157] (26);

[0158] After sorting, we can obtain:

[0159] (27);

[0160] According to the quadratic formula, we can obtain:

[0161] (28);

[0162] Solving for:

[0163] (29);

[0164] Substituting x1 and x2 into the expressions for y and z respectively, we get:

[0165] (30);

[0166] In summary:

[0167] (31).

[0168] However, since y1 = 0.1687 m exceeds the maximum geometric dimension of the transformer cavity, this solution is invalid. Therefore, according to the embodiment, the discharge positioning position of the three sensing components is calculated to be P(x,y,z) = (0.00199, -0.00394, 0.001). By comparing with the preset calibration discharge point (0,0,0), the actual positioning error is 4.5 mm, thus controlling the positioning error within millimeters.

[0169] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A method for diagnosing and locating discharge in transformer oil using a dual-mode electromagnetic-pressure approach, characterized in that, Includes the following steps: S1, based on three or more electromagnetic-pressure integrated sensing components installed in different positions inside the valve to be inspected, collects electromagnetic signals and pressure wave signals in real time. S2, determine the time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component and the time when the shock pressure wave arrives at each electromagnetic-pressure integrated sensing component, and then determine the distance from the discharge point to each electromagnetic-pressure integrated sensing component. S3. Based on the distance from the discharge point to each electromagnetic-pressure integrated sensing component, a set of positioning equations is constructed. S4. Solve the positioning equations to determine the location of the discharge point.

2. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 1, characterized in that, In step S1, the integrated electromagnetic-pressure sensing assembly includes a housing and a high-frequency current sensor and a piezoelectric high-frequency pressure sensor stacked together and installed inside the housing.

3. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 1, characterized in that, In step S2, the distance d from the discharge point to the i-th integrated electromagnetic-pressure sensing component is calculated according to the following formula. i : ； Among them, t ei t represents the time it takes for the electromagnetic wave to reach the i-th electromagnetic-pressure integrated sensing component; pi v is the time it takes for the shock pressure wave to reach the i-th electromagnetic-pressure integrated sensing component; e v is the speed at which electromagnetic waves propagate in transformer oil. p This refers to the propagation speed of the shock pressure wave in the transformer oil.

4. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 3, characterized in that, v p It will change with oil temperature, and the specific result can be determined using the following formula: ； Where α is the temperature correction factor; T is the oil temperature; This is a reference temperature.

5. The electromagnetic-pressure dual-mode fusion method for diagnosing and locating discharge in transformer oil according to any one of claims 1 to 4, characterized in that, Step S2 includes the following sub-steps: S21, Based on the set electromagnetic wave trigger threshold, determine the timing trigger start point of all electromagnetic-pressure integrated sensing components; S22, starting from the self-timer trigger point, within the adjustable monitoring window, determines the time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component based on the electromagnetic wave adaptive threshold. S23, starting from the self-timer trigger point, within the adjustable monitoring window, the time for the shock pressure wave to reach each electromagnetic-pressure integrated sensing component is determined based on the shock pressure wave adaptive threshold. S24. Based on the time it takes for the electromagnetic wave to reach each electromagnetic-pressure integrated sensing component and the time it takes for the impact pressure wave to reach each electromagnetic-pressure integrated sensing component, the distance from the discharge point to each electromagnetic-pressure integrated sensing component is calculated.

6. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 5, characterized in that, In step S22, the arrival time of electromagnetic waves at each electromagnetic-pressure integrated sensing component is determined according to the following steps: (A1) Calculate the time period T before the adjustable monitoring window or the start of the timing trigger. pre Internal electromagnetic signal V e,i The root mean square noise level N of (t) rms,e,i ; (A2) Based on electromagnetic signal V e,i The root mean square noise level N of (t) rms,e,i Construct an adaptive threshold V′ for electromagnetic waves th,e ; (A3) The time when the electromagnetic wave arrives at each electromagnetic-pressure integrated sensing component is the time when the corresponding signal amplitude first exceeds the adaptive threshold V′. th,e The moment t ei .

7. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 5, characterized in that, In step S23, the arrival time of the shock pressure wave at each electromagnetic-pressure integrated sensing component is determined according to the following steps: (B1) Calculate the time period T before the adjustable monitoring window or the start of the timing trigger. pre Internal shock pressure wave signal V p,i The root mean square noise level N of (t) rms,p,i ; (B2) Based on the impact pressure wave signal V p,i The root mean square noise level N of (t) rms,p,i Construct an adaptive threshold V for shock pressure waves th,p ; (B3) The time when the shock pressure wave arrives at each electromagnetic-pressure integrated sensing component is the time when the corresponding signal amplitude first exceeds the shock pressure wave adaptive threshold V. th,p The moment t pi .

8. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 1, characterized in that, In step S3, for n integrated electromagnetic-pressure sensing components, a set of n distance equations is constructed between the discharge point and the n integrated electromagnetic-pressure sensing components; Then, by using the finite difference method, the quadratic terms in each equation are eliminated, resulting in a system of n-1 linear equations.

9. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 8, characterized in that, The distance between the discharge point and the i-th electromagnetic-pressure integrated sensing component is defined as: ； Where (x,y,z) represents the location of the discharge point, (x i ,y i ,z i ) represents the position of the i-th electromagnetic-pressure integrated sensing component, and d i Let be the distance between the discharge point and the i-th electromagnetic-pressure integrated sensing component.

10. The method for diagnosing and locating discharge in transformer oil using electromagnetic-pressure dual-mode fusion as described in claim 9, characterized in that, The i-th integrated electromagnetic-pressure sensing component and the j-th integrated electromagnetic-pressure sensing component, after eliminating quadratic terms using the difference method, are expressed as follows: 。

Images