Frequency domain dielectric spectrum curve correction method and device and computer equipment

Abstract

The application relates to a frequency domain dielectric spectrum curve correction method and device, computer equipment, a storage medium and a computer program product. The frequency domain dielectric spectrum curve correction method comprises the following steps: obtaining first test data of a dielectric loss curve of a sleeve at different constant temperatures; obtaining a nonlinear relationship between a plurality of characteristic parameter values in a preset dielectric relaxation model and temperature according to the first test data; obtaining an equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the sleeve at a time-varying temperature according to the dielectric relaxation model and the nonlinear relationship; and performing translation correction on the frequency points of the frequency domain dielectric spectrum curve at the time-varying temperature according to the Arrhenius formula and the corresponding relationship between the test frequency points and the equivalent temperatures.

Description

Technical Field

[0001] This application relates to the field of oil-paper insulation technology, and in particular to a method, apparatus and computer equipment for correcting frequency domain dielectric spectrum curves. Background Technology

[0002] Frequency domain dielectric spectroscopy (FDS) technology is widely used in the condition diagnosis of oil-paper insulated electrical equipment. However, because the test time for frequency domain dielectric spectroscopy curves in the low-frequency band is relatively long, the temperature of the equipment is often dynamically changing during the test. Furthermore, existing evaluation databases are based on constant-temperature frequency domain dielectric spectroscopy curves, resulting in certain errors in the insulation condition assessment using time-varying temperature frequency domain dielectric spectroscopy curves. Summary of the Invention

[0003] Therefore, it is necessary to provide a correction method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can correct a time-varying temperature frequency domain dielectric spectrum curve to a frequency domain dielectric spectrum curve under constant temperature conditions, in order to address the above-mentioned technical problems.

[0004] A method for correcting frequency domain dielectric spectrum curves, comprising:

[0005] First test data of the dielectric loss curves of the bushing at different constant temperatures were obtained.

[0006] Based on the first test data, obtain the nonlinear relationship between multiple characteristic parameter values ​​and temperature in the preset dielectric relaxation model;

[0007] Based on the dielectric relaxation model and the nonlinear relationship, the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature is obtained;

[0008] Based on the Arrhenius formula and the correspondence between the test frequency and the equivalent temperature, the frequency points of the frequency domain dielectric spectrum curve under time-varying temperature are shifted and corrected.

[0009] In one embodiment, obtaining the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature based on the dielectric relaxation model and the nonlinear relationship includes:

[0010] Based on the dielectric relaxation model, the expressions for the real and imaginary parts of the complex permittivity are obtained respectively;

[0011] Based on the nonlinear relationship, the real and imaginary parts of the complex permittivity are solved using the least squares method to obtain the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature.

[0012] In one embodiment, before solving for the real and imaginary parts of the complex permittivity using the least squares method based on the nonlinear relationship, the method further includes:

[0013] Construct an equivalent scaled-down model of the casing;

[0014] Heat conduction analysis is performed based on the equivalent scaled-down model to obtain the heat dissipation equation of the bushing. The heat dissipation equation is used to characterize the relationship between heat dissipation time and the insulation temperature inside the bushing.

[0015] In one embodiment, the step of solving for the real and imaginary parts of the complex permittivity using the least squares method based on the nonlinear relationship includes:

[0016] The simulated temperature at the target time is obtained based on the heat dissipation equation.

[0017] The simulated temperature is used as the initial value for the least squares method, and the real and imaginary parts of the complex permittivity are solved using the least squares method based on the initial value.

[0018] In one embodiment, obtaining the nonlinear relationship between multiple characteristic parameter values ​​and temperature in a preset dielectric relaxation model based on the first test data includes:

[0019] Based on the first test data, multiple characteristic parameter values ​​in the dielectric relaxation model are obtained through a heuristic algorithm;

[0020] The nonlinear relationship between multiple characteristic parameters and temperature is obtained by using a cubic spline interpolation function.

[0021] In one embodiment, it further includes:

[0022] Obtain the second test data of the dielectric loss curve of the bushing under time-varying temperature;

[0023] The second test data is compared with the frequency domain dielectric spectrum curve after translation correction to verify the accuracy of the correction method.

[0024] A device for correcting frequency domain dielectric spectrum curves, comprising:

[0025] The data acquisition module is used to acquire the first test data of the dielectric loss curve of the bushing at different constant temperatures;

[0026] The nonlinear relationship calculation module is used to obtain the nonlinear relationship between multiple characteristic parameter values ​​and temperature in the preset dielectric relaxation model based on the first test data.

[0027] The equivalent temperature acquisition module is used to acquire the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature, based on the dielectric relaxation model and the nonlinear relationship.

[0028] The translation correction module is used to perform translation correction on the frequency points of the frequency domain dielectric spectrum curve under time-varying temperature according to the Arrhenius formula and the correspondence between the test frequency points and the equivalent temperature.

[0029] A computer device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method described above.

[0030] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0031] A computer program product includes a computer program that, when executed by a processor, implements the steps of the method described above.

[0032] The aforementioned method, apparatus, computer equipment, storage medium, and computer program product for correcting frequency domain dielectric spectrum curves include a method that, by providing a constant temperature environment and acquiring first test data at each constant temperature, trains a preset dielectric relaxation model to obtain specific values ​​for multiple characteristic parameters within the preset dielectric relaxation model, thereby acquiring an accurate dielectric relaxation model. Based on the dielectric relaxation model and nonlinear relationships, the equivalent temperature corresponding to each test frequency point can be obtained. By obtaining the equivalent temperature, the test frequency points of the data acquired under time-varying temperature conditions can be shifted, thus providing an accurate frequency domain dielectric spectrum curve under constant temperature conditions. Attached Figure Description

[0033] Figure 1 Here are the tanδ-f curves under constant temperature and time-varying temperature conditions in one embodiment;

[0034] Figure 2 This is one of the flowcharts for a method to correct a frequency domain dielectric spectrum curve according to an embodiment;

[0035] Figure 3 This is a schematic diagram of an experimental platform according to one embodiment;

[0036] Figure 4 Here are the tanδ-f curves at different constant temperatures in one embodiment;

[0037] Figure 5 Here are the C′-f curves at different constant temperatures in one embodiment;

[0038] Figure 6Here are the C″-f curves at different constant temperatures in one embodiment;

[0039] Figure 7 This is a schematic diagram of the particle movement rules in the PSO algorithm.

[0040] Figure 8 This is a temperature distribution cloud map of the bushing after 180 minutes of heat dissipation in one embodiment.

[0041] Figure 9 The figure shows the temperature change over time of the sleeve capacitor core in one embodiment.

[0042] Figure 10 This is a comparison graph of the sleeve temperature change over time in one embodiment and the simulation curve.

[0043] Figure 11 The frequency domain dielectric spectrum curve under time-varying temperature conditions is shown in one embodiment.

[0044] Figure 12 The calibration result is for a reference temperature of 30°C in one embodiment;

[0045] Figure 13 The calibration result is for a reference temperature of 60°C in one embodiment;

[0046] Figure 14 The calibration result is for a reference temperature of 90°C in one embodiment;

[0047] Figure 15 This is an internal structural diagram of a computer device according to one embodiment. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] In related technologies, to obtain accurate frequency-domain dielectric spectrum curves, frequency-domain dielectric spectroscopy can be used to study the insulation properties of different insulating oils at different constant temperatures, and the frequency dependence of their dielectric response can be characterized based on activation energy. Alternatively, by preparing samples with different degrees of moisture and aging, the influence of temperature on the low-frequency nonlinear conductivity loss characteristics of oil-paper insulation can be studied. Furthermore, the effects of moisture and temperature in the bushing on the frequency-domain dielectric spectrum curves can be discussed, and a curve correction method based on a compensation factor can be proposed to calibrate the frequency-domain dielectric spectrum curves at other steady-state temperatures to the reference temperature conditions.

[0050] However, the aforementioned studies are all based on frequency domain dielectric spectrum curve evaluation methods at steady-state temperatures. In actual operating conditions, the field environment is more complex, especially affected by external factors such as location and climate. Therefore, when obtaining dielectric response data during shutdown maintenance, the equipment is often in a dynamic cooling process, and the testing cycle for low-frequency bands is relatively long, with a complete testing process typically taking nearly one hour. Moreover, because the "time window" for on-site maintenance is short, the time for dielectric response testing is limited, making it impossible to wait for the bushing temperature to stabilize before evaluating the dielectric response. For these reasons, when conducting dielectric response testing under operating conditions, the difference between the starting and ending temperatures is significant, resulting in a large deviation between the test curve and the frequency domain dielectric spectrum curve at steady-state temperatures.

[0051] Specifically, Figure 1 Here are the tanδ-f curves under constant temperature and time-varying temperature conditions in one embodiment, with reference to... Figure 1 In high-frequency bands, the frequency domain dielectric spectrum curve has fewer test points, resulting in shorter test times and minimal temperature changes. Therefore, the difference between the time-varying temperature curve and the constant temperature curve is small at high frequencies. As the test frequency decreases, the temperature drops, and the time-varying temperature curve shifts to the left, gradually increasing the difference from the constant temperature curve. This is because during heat dissipation, the carrier mobility within the oil-paper insulation decreases significantly, leading to reduced conductivity losses at low frequencies and a leftward shift of the curve. Since the low-frequency band of the frequency domain dielectric spectrum curve is closely related to the moisture aging state of the oil-paper insulation, differences in the low-frequency band reduce the accuracy of insulation condition assessment results. Furthermore, because traditional frequency domain dielectric spectrum curve assessment databases are often built based on constant temperature curves, the time-varying temperature curve needs to be corrected to a constant temperature curve under operating conditions to achieve accurate assessment of the insulation condition of power equipment. Simultaneously, because traditional assessment databases are based on constant temperature curves, the time-varying temperature curve needs to be corrected to constant temperature conditions for effective insulation assessment.

[0052] Therefore, embodiments of this application provide a method for correcting frequency domain dielectric spectrum curves. Figure 2 Here is one of the flowcharts for a method to correct a frequency domain dielectric spectrum curve according to an embodiment, see reference. Figure 2 In this embodiment, the correction method for the frequency domain dielectric spectrum curve includes steps S100 to S400.

[0053] S100, to obtain the first test data of the dielectric loss curve of the bushing at different constant temperatures.

[0054] Figure 3 This is a schematic diagram of an experimental platform according to one embodiment, with reference to... Figure 3The experimental platform allows for the testing of the bushing's frequency domain dielectric spectrum curves at different constant temperatures, yielding initial test data on the bushing's dielectric loss curves at these temperatures. The moisture content of the bushing can be, for example, 0.56%. Specifically, the bushing can be placed vertically in a constant temperature and humidity oven, connected to a frequency domain dielectric spectroscopy (FDS) analyzer and a PC controller via an interface on the oven's outer shell. Electrodes embedded within the oven connect the high-voltage end and the test end to the bushing's guide rod and end screen, respectively, for testing. Based on this experimental platform, the PC controller can control the oven temperature and analyze the data output from the frequency domain dielectric spectroscopy analyzer to obtain the initial test data for the bushing.

[0055] For example, the test temperature may include, but is not limited to, 30°C, 40°C, 50°C, 60°C, 70°C, 80°C, and 90°C. Specifically, the temperature can be adjusted by gradient heating, and the temperature can be stabilized for 5 hours at each temperature before measurement. Figure 4 Here are the tanδ-f curves at different constant temperatures in one embodiment. Figure 5 Here are the C′-f curves at different constant temperatures in one embodiment. Figure 6 The C″-f curves are shown for different constant temperatures in one embodiment. (Reference) Figure 4 The tanδ-f curves at different temperatures show obvious intersections in the mid-to-high frequency bands. In the low-frequency portion of the tanδ-f curve below the intersection point, the dielectric loss increases with increasing temperature. In the high-frequency portion of the tanδ-f curve above the intersection point, the dielectric loss decreases with increasing temperature. Overall, the tanδ-f curves exhibit a shifting trend. (Reference) Figure 5 For the C′-f curve, a significant upward tilt appears in the low-frequency region with increasing temperature. This upward tilt is mainly due to two reasons: firstly, increased temperature reduces the interfacial polarization time constant between the oil and paper, allowing interfacial polarization to complete in a shorter time. Secondly, according to the low-frequency dispersion theory, recombination and dissociation tend to reach equilibrium during the oscillation of ions in the high-frequency band. As the frequency decreases, some ions are blocked and bound due to increased path length, leading to an increase in the real part of the complex capacitance. Temperature intensifies ion recombination and dissociation, making low-frequency dispersion more pronounced. (Reference) Figure 6 The C″-f curve reflects the conductivity and polarization losses of oil-paper insulation, and it exhibits the same trend as the tanδ-f curve. The low-frequency portion of the C″-f curve primarily reflects conductivity losses; due to increased temperature, the mobility of charge carriers within the oil-paper insulation increases significantly. The high-frequency portion of the C″-f curve primarily reflects polarization losses; increased temperature intensifies molecular thermal motion, making molecular polarization more difficult and thus reducing polarization losses.

[0056] S200: Based on the first test data, obtain the nonlinear relationship between multiple characteristic parameter values ​​and temperature in the preset dielectric relaxation model.

[0057] Among them, the values ​​of several characteristic parameters are different for different temperatures. Therefore, by obtaining the values ​​of several characteristic parameters corresponding to each temperature, the accuracy of subsequent calculations can be effectively improved. Specifically, for the frequency domain dielectric spectrum curves at different temperatures, the values ​​of εs, τ, α, and β are all different. Among them, εs can be expressed as a function of the micro-polarizability ae of electronic polarization and the micro-polarizability a0 of dipole polarization. Since the electronic structure in an atom is independent of temperature, ae is not affected by temperature. However, the energy of molecules is different at different temperatures, and their orientation polarization is different, so a0 is related to temperature. Moreover, the relaxation of the medium is caused by the energy transfer between particles in the system. That is, the higher the temperature, the faster the energy transfer speed, and the smaller the relaxation time τ. And the energy of the particles is distributed according to Boltzmann, so the relaxation time τ has an exponential relationship with temperature. Among them, α and β are shape parameters and have no physical meaning. Thus, the functional relationship between the dielectric loss value and temperature T can be derived as shown in equations (14) and (15). Where N is the number of independent permanent dipoles, R is the relevant atomic radius, ε0 is the vacuum permittivity, A is a constant coefficient, U is the molecular activation energy, and k is the Boltzmann constant, k = 1.38 × 10⁻²³ J / K.

[0058] tanδ=f(ε s (T), ε ∞ ,τ(T),α(T),β(T)) (14)

[0059]

[0060] S300, based on the dielectric relaxation model and nonlinear relationship, obtains the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature.

[0061] Under time-varying temperature conditions, the dielectric spectrum curves in the low-frequency band (0.1Hz~1mHz) require a long testing time, and the temperature at the corresponding test frequency points is not a constant. Therefore, by obtaining the equivalent temperature in this step, accurate temperature correction of the curve can be performed in subsequent steps.

[0062] S400, based on the Arrhenius formula and the correspondence between the test frequency and the equivalent temperature, performs frequency shift correction on the frequency domain dielectric spectrum curve under time-varying temperature.

[0063] The Arrhenius equation, as shown in equation (17), is used to shift a frequency point under a time-varying temperature to the desired constant temperature condition. Here, Ts is the target temperature for correction; f0 is the frequency of a point on the frequency domain dielectric spectrum curve at the equivalent temperature Tequal before shifting; f is the frequency of that point at temperature Ts after shifting; Ea is the activation energy of the insulating paper; and k is the Boltzmann constant.

[0064]

[0065] In this embodiment, by providing a constant temperature environment and acquiring first test data at each constant temperature in the above-mentioned steps, a preset dielectric relaxation model can be trained to obtain the specific values ​​of multiple characteristic parameters in the preset dielectric relaxation model, thereby obtaining an accurate dielectric relaxation model. Based on the dielectric relaxation model and nonlinear relationship, the equivalent temperature corresponding to each test frequency point can be obtained. By obtaining the equivalent temperature, the test frequency points of the data acquired under time-varying temperature conditions can be shifted, thereby providing an accurate frequency domain dielectric spectrum curve under constant temperature conditions.

[0066] In one embodiment, step S200 obtains the nonlinear relationship between multiple characteristic parameter values ​​and temperature in a preset dielectric relaxation model based on the first test data, including steps S210 to S220.

[0067] S210: Based on the first test data, obtain the values ​​of multiple characteristic parameters in the dielectric relaxation model through a heuristic algorithm.

[0068] In classical dielectric theory, the Dybe model with a single relaxation time is commonly used to describe dielectric processes. However, in most cases, the relaxation time follows a maximum probability distribution. Therefore, in this embodiment, the dielectric relaxation model can be the HN (Havriliak-Negami) model. The HN model is a combination of the Cole-Cole equation and the Cole-Davidson function, providing a more general function for explaining the dielectric and mechanical relaxation processes of certain polymer systems. The expression of the HN model is shown in equation (9). Where j is the unit imaginary number; τ is the relaxation time constant; εs and ε∞ represent the static dielectric constant and the optical frequency dielectric constant, respectively; α and β are shape parameters related to the relaxation time distribution, 0≤α≤1, 0≤β≤1.

[0069]

[0070] However, the HN model has several drawbacks. First, its expression is quite complex. Solving the relationship between its parameters and temperature using traditional methods is an NP-hard problem, characterized by discretization, multiple indices, nonlinearity, and uncertainty. Furthermore, the time and space costs of solving multiple parameters are enormous. Second, traditional methods require high-quality initial values ​​for parameter fitting; initial values ​​that deviate significantly from the optimal solution can lead to model performance degradation and increased error in the solution.

[0071] Therefore, in this embodiment, the heuristic algorithm PSO (Particle Swarm Optimization) is used to solve for the parameters of the HN model under different constant temperatures. The advantages of the PSO algorithm are its simple algorithm flow, ease of implementation, and relatively few parameters to be adjusted. During the particle swarm optimization process, by adjusting the relation weight coefficients, self-learning factors, and social learning factors, good development capabilities can be obtained in the early stages of optimization, tending towards global search and achieving good search capabilities. In the later stages, it tends towards local search, thereby improving the accuracy of the solution. To ensure the accuracy of the results, in this embodiment, the objective function is to minimize the mean square error between the calculated complex permittivity and the measured complex permittivity, and the solution is the HN model parameters, transforming the problem into an optimization problem.

[0072] Considering the characteristics of the HN model, the parameter settings of the algorithm can be as shown in Table 2.

[0073] Table 2 PSO Algorithm Parameters

[0074]

[0075] Specifically, the PSO algorithm process is as follows. A particle swarm consists of n particles within the feasible region, and the position of each particle represents a feasible solution. The position of the i-th particle after the (k+1)-th iteration can be represented as shown in equation (18). The selection of parameters such as particle self-learning factor c1, social learning factor c2, and maximum iteration number ngen will affect the convergence speed of the PSO algorithm.

[0076]

[0077] The update of feasible solutions is accomplished through particle movement. The position of the i-th particle after the (k+1)-th iteration in dimension d is given by equation (19). Here, x is the parameter of the HN model, and v is the velocity of the particle swarm. The velocity of the particle swarm can be expressed as shown in equation (20). Here, ω is the inertia weight; c1 is the self-learning factor, c2 is the social learning factor; r1 and r2 are random numbers on [0,1] to increase the randomness of the search; pb is the historical best position of particle i; and gb is the historical best position of the entire particle swarm. Figure 7This is a schematic diagram of the particle movement rules in the PSO algorithm, for reference. Figure 7 The particle's movement vector consists of three parts: the particle's original movement direction, the difference between the particle's current position and its historical best position, and the difference between the particle's current position and the population's best position. In the PSO algorithm, the quality of the solution is measured by particle fitness, and the particle fitness function is defined as shown in equation (21). In this equation, the subscript pso represents the calculated value obtained by substituting the optimal result of the particle swarm optimization algorithm into equations (10) and (11), test represents the measured value of the frequency domain dielectric spectrum curve, and s represents the test frequency point of the frequency domain dielectric spectrum curve.

[0078]

[0079]

[0080]

[0081] Based on equations (14) and (15), the corresponding five parameter values ​​in the HN model can be retrieved using frequency domain dielectric spectrum curve data at different constant temperatures, as shown in Table 3. Table 3 shows multiple characteristic parameter values ​​corresponding to temperatures of 30℃, 40℃, 50℃, 60℃, 70℃, 80℃, and 90℃.

[0082] Table 3 HN model parameters at different temperatures

[0083]

[0084] S220 employs cubic spline interpolation to obtain the nonlinear relationship between multiple characteristic parameters and temperature. Since it is necessary to correct for multiple time-varying temperatures between 30℃ and 90℃, cubic spline interpolation can be achieved through a smooth curve derived from a series of shape points, and mathematically, this is accomplished by solving a system of three bending moment equations to obtain the curve function set. Therefore, cubic spline interpolation is used to construct the accurate relationship between the parameters of the HN model and temperature.

[0085] In this embodiment, the PSO algorithm is used to solve the relationship between HN model parameters and temperature. This can eliminate problems such as discretization, nonlinearity and uncertainty in the solution process caused by the complexity of the model expression, reduce the time and space costs of computation, and improve the solution accuracy.

[0086] In one embodiment, step S300 obtains the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature based on the dielectric relaxation model and nonlinear relationship, including steps S310 to S320.

[0087] Step S310: Based on the dielectric relaxation model, obtain the expressions for the real and imaginary parts of the complex permittivity. According to complex analysis theory, the expressions for the real and imaginary parts of the complex permittivity can be obtained from equation (9) as equations (10) and (11), respectively.

[0088]

[0089]

[0090] S320, based on the nonlinear relationship, solves for the real and imaginary parts of the complex permittivity using the least squares method to obtain the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature.

[0091] Specifically, based on equations (10) and (11), the expression for dielectric loss can be obtained as shown in equation (12), where the expression for θ is shown in equation (13). Furthermore, by combining equations (14) and (15), the corresponding equivalent temperature can be obtained as shown in equation (16) when the frequency of a certain point in the dielectric spectrum and the dielectric loss value are known. However, it can be seen from the reference that equation (16) is quite complex and its explicit expression cannot be derived. Therefore, in this embodiment, the least squares (LS) method is used for solving. The least squares method is a mathematical optimization technique. It finds the best function match for the data by minimizing the sum of squares of the errors. Using the least squares method, the unknown parameters of the HN model can be easily obtained, minimizing the sum of squares of the errors between these parameters and the actual data.

[0092]

[0093]

[0094] T equal =f -1 (tanδ,ω) (16)

[0095] In one embodiment, before step S320 solves for the real and imaginary parts of the complex permittivity using the least squares method based on the nonlinear relationship, steps S330 to S340 are also included.

[0096] S330, construct an equivalent scaled-down model of the bushing.

[0097] The simulation model uses the same structure and materials as the actual bushing. The structure of the simulation model is determined based on the central current-carrying conductor, insulating paper, aluminum foil, transformer oil, outer sheath, and air gap. Its main insulation can be equivalent to a coaxial series capacitor composed of multiple layers of aluminum foil electrodes. The central current-carrying conductor is a copper conductor with high thermal conductivity, and the outer shell is made of polymethyl methacrylate (PMMA) acrylic glass. The bushing model parameters are shown in Table 4.

[0098] Table 4 Parameters of the Casing Model

[0099] Number of electrode layers Thickness (mm) Upper range (mm) Lower range (mm) Plate length (mm) 0 / / / 260 1 1.6 24 6 220 2 1.6 29 8 180 3 1.6 34 8 140 4 1.6 57 16 65

[0100] The construction of an equivalent scaled-down model of the bushing includes the following steps: A two-dimensional axisymmetric model of the bushing is constructed, and the model is meshed. Parameters such as constant-pressure heat capacity, thermal conductivity, and density are set for materials such as the guide rod, transformer oil, insulating paper, and aluminum foil. A transient solver and simulation time are designed, and domain probes are set to obtain two-dimensional and three-dimensional temperature distribution cloud maps of the bushing. The radius R0 of the zero screen is 16.5 mm, and the length L0 is 260 mm. Simultaneously, the capacitor-type oil-paper bushing model structure includes solid, liquid, and gaseous states. Based on actual bushing operating conditions, the initial oil temperature is set to 70℃, the initial capacitor core temperature to 85℃, and the initial temperatures of the air gap and outer sheath to be the same as the ambient temperature. By constructing an equivalent scaled-down model of the bushing, temperature field simulation can be performed, thereby obtaining accurate temperature analysis results.

[0101] S330, heat conduction analysis is performed based on the equivalent scale model to obtain the heat dissipation equation of the bushing. The heat dissipation equation is used to characterize the relationship between heat dissipation time and the insulation temperature inside the bushing.

[0102] Specifically, the bushing is constantly subjected to power frequency AC voltage during operation, and its overall temperature remains constant. During power outages and maintenance, the bushing temperature changes due to heat conduction between the media, heat convection between the air and the bushing, and heat radiation from the bushing itself. However, the temperature obtained under operating conditions is often the bushing oil temperature, which cannot accurately reflect the bushing insulation temperature. That is, using the internal insulation temperature of the power equipment instead of the oil temperature for temperature correction yields more accurate results. Therefore, this embodiment uses COMSOL finite element simulation to analyze the heat dissipation process inside the bushing. Considering the bushing's heat dissipation environment, this embodiment mainly considers three heat dissipation processes: conductive heat dissipation, convection heat dissipation, and radiation heat dissipation.

[0103] Conductive heat dissipation refers to the heat transfer phenomenon in solid, liquid, and gaseous media when there is no macroscopic motion of molecules. Its differential equations are shown in equations (1) and (2). Where T is temperature in K, x, y, and z are coordinate values ​​in m, k is thermal conductivity in W·m⁻¹·K⁻¹, Cp is constant-pressure heat capacity in J·kg⁻¹·K⁻¹, ρ is density in kg·m⁻³, t is time in seconds, and Q is the heat output per unit volume in W·m⁻³. The material parameters during the simulation are shown in Table 5.

[0104]

[0105]

[0106] Table 5 Material heat dissipation parameters

[0107]

[0108] Convection heat dissipation refers to the convective heat transfer that occurs between the surface of the medium and the fluid. For capacitor-type oil-paper bushings, the outermost layer is in contact with the air. Due to the temperature difference in the vertical direction, the buoyancy of the air causes the gas to flow and conduct convective heat transfer, which is natural convection heat transfer. Transformer oil is sealed in the bushing and undergoes natural convection heat transfer with the bushing in the sealed cavity. The basic equations of thermal convection are as follows: mass conservation equation (3), momentum conservation equation (4), and energy conservation equation (5). Wherein, ν is the velocity, in m / s. F is the gravity acting on the fluid, in N. p is the fluid pressure, in Pa. η is the dynamic viscosity, in kg·m-1.

[0109]

[0110]

[0111]

[0112] The heat loss characteristics of the outer surface when dissipating heat through natural convection are usually characterized by the convective heat transfer coefficient h, as shown in equation (6). Here, L is the object size in meters (m). RaL and Pr are dimensionless Rayleigh and Prandtl numbers, respectively.

[0113]

[0114] Radiation-induced heat dissipation refers to the process where the surface of the casing also dissipates heat by radiating electromagnetic energy outwards. Heat is radiated from the heating element to the surrounding medium at a lower temperature in the form of waves. The magnitude of the radiative energy is related to the casing temperature and its surface properties. The maximum radiative flux density on the surface of an object during thermal radiation is given by equation (7). Where Ts is the absolute temperature of the object's surface, in K. σ is the Stefan-Boltzmann constant (σ = 5.67 × 10⁻⁸).

[0115]

[0116] In this embodiment, the simulation model considers the conductive heat dissipation inside the capacitor core, the external natural convection heat dissipation of the vertical and horizontal walls, and the radiative heat dissipation of the outer sheath surface to the environment, and obtains the temperature distribution cloud map of the sleeve under different heat dissipation times. Figure 8 This is a temperature distribution cloud diagram of the bushing after 180 minutes of heat dissipation in one embodiment, for reference. Figure 8The highest temperature in the bushing is located at the capacitor core, and the lowest temperature is located at the apex of the outer sheath. As the heat dissipation time increases, the internal temperature of the bushing gradually decreases, and the radial and axial temperatures of the capacitor core remain uniformly distributed throughout the heat dissipation process.

[0117] Figure 9 This is a curve showing the temperature change of the capacitor core over time in one embodiment. Figure 9 This provides a more intuitive understanding of the relationship between the internal insulation temperature of the bushing and heat dissipation time. Taking the bushing capacitor core as the research object, its heat dissipation curve is obtained. From... Figure 9 It can be seen that as the heat dissipation time increases, the temperature of the insulation inside the bushing does not decrease at a constant rate, but rather exhibits an exponential decreasing trend. The highest temperature is the initial temperature, and the lowest temperature is the ambient temperature, with the heat dissipation rate decreasing from fast to slow. The heat dissipation equation of the bushing can be obtained by fitting, as shown in equation (8). Thus, the insulation temperature at any moment during the heat dissipation process of the bushing can be obtained. Compared with using the oil temperature, time-varying temperature frequency domain dielectric spectrum curve correction based on this temperature will be more accurate.

[0118] T = 57.42e -0.00884t +25.82 (8)

[0119] In one embodiment, step S320 involves solving for the real and imaginary parts of the complex permittivity using the least squares method based on the nonlinear relationship, including steps S321 to S322.

[0120] S321, obtain the simulation temperature at the target time according to the heat dissipation equation. That is, obtain the simulation temperature at the target time through the heat dissipation equation of equation (8).

[0121] S322 uses the simulated temperature as the initial value for the least squares method, and solves for the real and imaginary parts of the complex permittivity based on the initial value using the least squares method.

[0122] Specifically, based on the aforementioned steps, the five characteristic parameter values ​​can be replaced with functions of temperature and substituted into equations (10)-(11) to solve for the equivalent temperature. Since equation (16) cannot be expressed as an explicit expression, and the known data exceeds the number of parameters to be solved, it is understandable that the least squares method heavily relies on the initial values, and the selection of the initial values ​​will have a significant impact on the final results. Therefore, using the simulated internal insulation temperature as the initial value for the least squares method through equation (8) can ensure the accuracy of the calculation results.

[0123] In one embodiment, the method for correcting the frequency domain dielectric spectrum curve further includes steps S510 to S520.

[0124] S510, acquire the second test data of the dielectric loss curve of the bushing under time-varying temperature.

[0125] Specifically, the frequency domain dielectric spectrum curve of an equivalent scaled-down model of the bushing was measured under time-varying temperatures in the laboratory. In the experiment, the bushing was first placed in a constant-temperature oven and heated to 85°C for 5 hours. Immediately afterwards, the bushing was removed for frequency domain dielectric spectrum testing. Two sets of T-type thermocouples were placed at the capacitor core conductor and in the surrounding transformer oil, respectively, to monitor the temperature change of the bushing during heat dissipation. Based on the second set of test data, the temperature change curve of the bushing over time could be obtained.

[0126] S520 compares the second test data with the shifted and corrected frequency domain dielectric spectrum curve to verify the accuracy of the correction method.

[0127] Specifically, Figure 10 This is a comparison graph of the sleeve temperature change over time in one embodiment and the simulation curve, for reference. Figure 10 As can be seen, the shapes and corresponding values ​​of the two curves are basically the same, but the experimental temperature is slightly higher than the simulated temperature. This is because the simulation result is the temperature of the capacitor core in the bushing, while the internal insulation temperature cannot be measured during the experiment. Therefore, the average value of the conductor temperature and oil temperature is used as a substitute, and the overall temperature is slightly higher than the internal insulation temperature. In addition, as the heat dissipation time increases, the experimental and simulated temperatures do not decrease at a constant rate, but rather show an exponential decreasing trend. The highest temperature is the initial temperature, and the lowest temperature is the ambient temperature. The heat dissipation rate shows a trend of decreasing from fast to slow. Based on the root mean square error, the similarity between the two curves is calculated to be 2.8909, which verifies the accuracy of the simulation model and the heat dissipation characteristics of the bushing.

[0128] In the experiment of this embodiment, the time-varying temperature frequency domain dielectric spectrum curves of the sleeve were tested after 25 min (Case 1), 84 min (Case 2), and 143 min (Case 3) of heat dissipation. Figure 11 Here is a frequency domain dielectric spectrum curve under time-varying temperature conditions according to one embodiment, with reference to... Figure 11 The test results for the mid-to-high frequency range (5kHz to 0.1Hz) are basically consistent with the constant temperature curve. As the test temperature decreases, the frequency domain dielectric spectrum curves below 0.1Hz in all three cases show a tendency to shift to the left, differing significantly from the constant temperature curve data. This is because the test period for the frequency domain dielectric spectrum curve at higher frequencies is shorter, resulting in smaller temperature differences and tanδ values ​​close to the measured values ​​at constant temperature. As the test frequency decreases, the electric field change period at the corresponding frequency points gradually increases. In the low-frequency band of the frequency domain dielectric spectrum curve, the bushing insulation temperature decreases exponentially, causing the tanδ value to shift to the left.

[0129] It is understandable that the dielectric loss in the low-frequency band mainly consists of two parts: conductivity loss and polarization loss. On the one hand, the dielectric conductivity current caused by the carrier mobility in the oil-paper insulation decreases as the conductivity decreases with decreasing temperature, and the corresponding conductivity loss also shows a downward trend. On the other hand, as can be seen from equation (8), the thermal motion of polar molecules is weaker at lower temperatures, the relaxation time τ of the insulating medium continuously increases, the contribution of relaxation polarization to the total loss gradually decreases, and the frequency domain dielectric spectrum curve shows a leftward shift trend. Therefore, in order to calibrate the time-varying temperature curves in the three cases to an arbitrary constant reference temperature, the equivalent temperature of each frequency point in the low-frequency band (0.1Hz~1mHz) can be calculated using the least squares method based on equation (16), where the initial temperature used in the least squares calculation can be determined according to the formula. Figure 8 The simulation results are shown below. The equivalent temperature (°C) at each frequency point is shown in Table 5.

[0130] Table 1. Equivalent temperature at each frequency point under time-varying temperature conditions.

[0131] (f / Hz) 0.001 0.002 0.005 0.01 0.022 0.046 0.1 Case 1 66.184 68.417 70.514 71.853 75.842 78.382 80.195 Case 2 50.811 52.568 54.004 54.708 57.052 58.317 59.019 Case 3 37.672 38.243 40.601 42.212 43.992 44.45 44.491

[0132] In this embodiment, 30℃, 60℃, and 90℃ were used as reference temperatures, and the time-varying temperature curves under the three conditions were corrected based on formula (17), where the activation energy was an empirical value of 0.98 eV. The correction results were compared with the frequency domain dielectric spectrum curve test results. Figures 12 to 14 As shown. Figure 12 The calibration result is shown for an example with a reference temperature of 30°C. Figure 13 The calibration result is based on a reference temperature of 60°C in one embodiment. Figure 14 The calibration result is based on a reference temperature of 90°C in one embodiment, in conjunction with the reference. Figures 12 to 14 It can be seen that, compared with the test results under the reference temperature conditions of 30℃, 60℃, and 90℃, the test results under the time-varying temperature conditions are distorted in the low-frequency part, and the curve shows a leftward shift trend. The correction results under the three conditions are basically consistent with the test results under the reference temperatures of 30℃, 60℃, and 90℃, and the curves overlap well, which verifies the accuracy of the correction model.

[0133] Therefore, it can be confirmed that the proposed correction method based on the HN model can effectively correct the test results under time-varying temperature conditions, thereby achieving accurate assessment of the insulation status of power equipment under time-varying temperature conditions. The results show that the similarity between the bushing's temperature curve over time and the simulated curve is 2.8909, verifying the accuracy of the simulation. Moreover, under time-varying temperature conditions, the test results in the high-frequency band (5kHz–0.1Hz) of the frequency domain dielectric spectrum curve are consistent with the reference temperature curve. As the test temperature decreases, the low-frequency band (0.1Hz–1mHz) of the frequency domain dielectric spectrum curve shows a leftward shift, with a greater difference compared to the constant temperature curve data. That is, the time-varying temperature curve correction method based on the HN model can effectively correct the test results under time-varying temperature conditions.

[0134] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated in this embodiment, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps.

[0135] Based on the same inventive concept, this application also provides a frequency domain dielectric spectrum curve correction device for implementing the frequency domain dielectric spectrum curve correction method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more frequency domain dielectric spectrum curve correction device embodiments provided below can be found in the limitations of the frequency domain dielectric spectrum curve correction method described above, and will not be repeated here. Specifically, the frequency domain dielectric spectrum curve correction device includes a data acquisition module, a nonlinear relationship calculation module, an equivalent temperature acquisition module, and a translation correction module.

[0136] The system comprises the following modules: a data acquisition module for acquiring first test data of the dielectric loss curves of the bushing at different constant temperatures; a nonlinear relationship calculation module for obtaining the nonlinear relationship between multiple characteristic parameter values ​​in a preset dielectric relaxation model and temperature based on the first test data; an equivalent temperature acquisition module for obtaining the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperatures based on the dielectric relaxation model and the nonlinear relationship; and a translation correction module for performing translation correction on the frequency points of the frequency domain dielectric spectrum curve under time-varying temperatures based on the Arrhenius formula and the correspondence between test frequencies and equivalent temperatures.

[0137] Each module in the aforementioned frequency domain dielectric spectrum correction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0138] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 15 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for correcting a frequency domain dielectric spectrum curve. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0139] Those skilled in the art will understand that Figure 15 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0140] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the methods described in the foregoing embodiments.

[0141] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the methods described in the foregoing embodiments.

[0142] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the methods described in the foregoing embodiments.

[0143] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0144] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0145] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for correcting frequency domain dielectric spectrum curves, characterized in that, include: First test data of the dielectric loss curves of the bushing at different constant temperatures were obtained. Based on the first test data, multiple characteristic parameter values ​​in the dielectric relaxation model are obtained through a heuristic algorithm; The nonlinear relationship between multiple characteristic parameters and temperature is obtained by using a cubic spline interpolation function; Several characteristic parameters include the micro-polarizability of electronic polarization and the circumpolarizability of dipole polarization as functions of es, relaxation time t, shape parameter a, and shape parameter b; Based on the dielectric relaxation model, the expressions for the real and imaginary parts of the complex permittivity are obtained respectively; Construct an equivalent scaled-down model of the casing; Heat conduction analysis is performed based on the equivalent scaled-down model to obtain the heat dissipation equation of the bushing. The heat dissipation equation is used to characterize the relationship between heat dissipation time and the insulation temperature inside the bushing. The simulated temperature at the target time is obtained based on the heat dissipation equation. The simulated temperature is used as the initial value for the least squares method, and the real and imaginary parts of the complex permittivity are solved by the least squares method based on the initial value to obtain the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature. Based on the Arrhenius formula and the correspondence between the test frequency and the equivalent temperature, the frequency points of the frequency domain dielectric spectrum curve under time-varying temperature are shifted and corrected.

2. The method for correcting the frequency domain dielectric spectrum curve according to claim 1, characterized in that, Also includes: Obtain the second test data of the dielectric loss curve of the bushing under time-varying temperature; The second test data is compared with the frequency domain dielectric spectrum curve after translation correction to verify the accuracy of the correction method.

3. The method for correcting the frequency domain dielectric spectrum curve according to claim 1, characterized in that, The dielectric relaxation model includes the Havriliak-Negami model.

4. The method for correcting the frequency domain dielectric spectrum curve according to claim 3, characterized in that, The heuristic algorithm includes the particle swarm optimization algorithm, PSO. The method of obtaining multiple characteristic parameter values ​​in the dielectric relaxation model through heuristic algorithms includes: The particle swarm optimization algorithm (PSO) was used to obtain the values ​​of multiple feature parameters in the Havriliak-Negami model.

5. A device for correcting frequency domain dielectric spectrum curves, characterized in that, include: The data acquisition module is used to acquire the first test data of the dielectric loss curve of the bushing at different constant temperatures; The nonlinear relational operation module is used to obtain the values ​​of multiple characteristic parameters in the dielectric relaxation model based on the first test data using a heuristic algorithm. The nonlinear relationship between multiple characteristic parameters and temperature is obtained by using a cubic spline interpolation function; Several characteristic parameters include the micro-polarizability of electronic polarization and the circumpolarizability of dipole polarization as functions of es, relaxation time t, shape parameter a, and shape parameter b; The equivalent temperature acquisition module is used to obtain the expressions for the real part and imaginary part of the complex dielectric constant according to the dielectric relaxation model; construct the equivalent scaled model of the bushing; and perform heat conduction analysis according to the equivalent scaled model to obtain the heat dissipation equation of the bushing, which is used to characterize the relationship between heat dissipation time and the insulation temperature inside the bushing. The simulated temperature at the target time is obtained according to the heat dissipation equation; the simulated temperature is used as the initial value of the least squares method, and the real and imaginary parts of the complex permittivity are solved by the least squares method based on the initial value to obtain the equivalent temperature corresponding to each test frequency point of the dielectric loss curve of the bushing under time-varying temperature. The translation correction module is used to perform translation correction on the frequency points of the frequency domain dielectric spectrum curve under time-varying temperature according to the Arrhenius formula and the correspondence between the test frequency points and the equivalent temperature.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images