A simulation analysis method for transmission line flashover tripping

Abstract

The present invention proposes a simulation analysis method for flashover tripping of transmission lines. By establishing a mathematical model, the current, voltage waveforms and fault duration under different short-circuit conditions and shock wave parameters are predicted. Combined with the data collected by the sound sensor, the simulation model is used to obtain the sound field distribution and time domain changes. Finally, by analyzing the sound field model, the distribution and time domain changes of sound near the transmission line can be accurately predicted, providing data support for the layout of air micro-vibration sensors and fault point location. This method explores the mechanism of flashover tripping caused by external force damage through theoretical analysis, and establishes the intrinsic connection between short-circuit conditions and shock wave parameters. At the same time, according to the sound signal data collected during the short circuit, multi-physical field finite element simulation modeling is carried out to obtain the distribution and time domain changes of the sound field, thereby realizing the analysis and intelligent diagnosis of flashover tripping of transmission lines, providing important support for the safe and stable operation of the power industry.

Description

Technical Field

[0001] The invention relates to the technical field of power transmission line fault detection, and in particular to a simulation analysis method for power line flashover tripping. Background Art

[0002] The electric power industry is entering an era of large-scale construction. However, this development trend is also accompanied by a sharp increase in electricity consumption, which in turn increases the demand for power generation and transmission lines. Although transmission lines are usually able to reliably transmit power during operation, failures may occasionally occur, which may be caused by defects in the equipment itself, damage from external forces, or force majeure. Common forms of failure include lightning tripping and external force damage tripping.

[0003] In order to ensure that timely perception and appropriate measures can be taken when a transmission line fails, it is crucial to evaluate the health of the pins at the connection points of the transmission line. However, the existing methods for detecting flashover tripping of transmission lines are mainly based on a rough judgment of the fault and a comprehensive analysis. These analysis methods usually need to consider multiple factors such as weather conditions and protection action conditions, and often require a lot of financial and material resources. Staff need to monitor weather conditions in real time and detect whether the circuit is short-circuited. At the same time, these methods often cannot accurately determine the specific cause of the flashover tripping of the transmission line, nor can they accurately predict the fault, thus affecting the guarantee of the safety and stability of the power system. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a simulation analysis method for transmission line flashover tripping, which is aimed at the prediction and detection of transmission line flashover tripping, thereby enabling comprehensive and effective evaluation of transmission line flashover tripping.

[0005] A simulation analysis method for transmission line flashover tripping, characterized in that it comprises the following steps:

[0006] S1, constructing a mathematical short-circuit model of a transmission line, wherein the mathematical short-circuit model of a transmission line is formed by constructing a relationship between a shock wave parameter and a short-circuit condition, thereby obtaining current and voltage waveforms and fault duration of the transmission line under different short-circuit conditions and shock wave parameters;

[0007] S2, comprehensively analyze the current, voltage waveforms and fault duration of the transmission line under different short-circuit conditions and shock wave parameters obtained through the mathematical short-circuit model of the transmission line and the experimental data, so as to obtain the short-circuit conditions and shock wave size when the transmission line trips;

[0008] S3, predicting the short-circuit condition of the transmission line through a simulation model under the short-circuit condition, thereby obtaining prediction data, wherein the prediction data is the tripping probability under different short-circuit conditions;

[0009] S4, obtaining acoustic signal data generated during a short circuit, and inputting the acoustic signal data and the predicted data into a convolutional neural network to obtain feature data, wherein the acoustic signal data is obtained by arranging an acoustic sensor near the transmission line;

[0010] S5, inputting the characteristic data into a multi-physics field finite element simulation model, thereby obtaining a sound field distribution model and a time domain change of the sound field in the vicinity of the transmission line short circuit;

[0011] S6, visualizes or outputs the acoustic field distribution near the short circuit of the transmission line and the time domain variation results, and performs status monitoring and fault diagnosis of the transmission line.

[0012] Furthermore, the shock wave parameters in step S1 include shock voltage and shock current.

[0013] Furthermore, the specific calculation formula of the transmission line mathematical short-circuit model is: Where g(t) represents the instantaneous conductance of the arc, g(t+Vt) represents the value of the instantaneous conductance of the arc after a period of time step vT, [Gg(t)] represents the difference between the steady-state conductance G of the arc and the instantaneous conductance g(t), The correction term represents the decay trend of the arc dynamic conductivity over time.

[0014] Furthermore, the experimental data in step S2 is obtained by placing a high-speed camera and a current and voltage sensor near the transmission line to record the changes in short-circuit voltage, current waveform, shock wave parameters and transmission line behavior response data during the short-circuit process.

[0015] Furthermore, the step S4 preprocesses the acoustic signal data and the prediction data before inputting them into the convolutional neural network, and the preprocessing includes normalization, data set downloading and loading steps.

[0016] Furthermore, it is characterized in that the step S5 comprises the following steps:

[0017] S51, establishing a multi-physics field finite element simulation model, wherein the multi-physics field finite element simulation model uses finite element software to establish three physical fields: electric field, thermal field and acoustic field;

[0018] S52, performing electrothermal coupling and thermoacoustic coupling settings on the multi-physics field finite element simulation model;

[0019] S53, based on the multi-physics field model and coupling setting, deriving the acoustic wave signal model and discretizing it;

[0020] S54, obtaining the spatial distribution model and time domain changes of the sound field through the sound wave signal model after discrete processing.

[0021] Furthermore, the specific calculation formula of the acoustic wave signal model is: Among them, u is the sound wave signal, t is time, x is the spatial coordinate, v is the speed of sound, and f is the external excitation.

[0022] Furthermore, the specific calculation formula of the acoustic wave signal model after discrete processing is (M 2 +v 2 Δt 2 ) n+1 =M(2U n -U n-1 )+Δt 2 F, where M is the mass matrix, v is the speed of sound, Δt is the time step, and U n+1 is the unknown displacement vector of the next time step, U n and U n-1 are the known displacement vectors of the current and previous time steps respectively, and the F vector represents the source term.

[0023] The present invention provides a simulation analysis method for transmission line flashover tripping, which can accurately predict the short-circuit conditions and shock wave size when the transmission line trips, thereby overcoming the limitations of the prior art of rough judgment and comprehensive analysis methods of faults, achieving accurate prediction of faults, improving the safety and stability of the power system, and reducing the problem of consuming a large amount of financial and material resources, providing important technical support for the development of the power industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of the steps of extracting characteristic sound signals by a convolutional neural network in a simulation analysis method for flashover tripping of a transmission line proposed by the present invention;

[0025] Figure 2 A multi-physics field simulation technology roadmap for flashover phenomenon in a simulation analysis method for flashover tripping of a transmission line proposed by the present invention;

[0026] Figure 3 A transmission line flashover tripping analysis route map in a transmission line flashover tripping simulation analysis method proposed by the present invention;

[0027] Figure 4 Reading the MN I ST data set code in a simulation analysis method for flashover tripping of a transmission line proposed by the present invention;

[0028] Figure 5A schematic diagram of the structure of a CNN model in a simulation analysis method for flashover tripping of a transmission line proposed by the present invention;

[0029] Figure 6 A schematic diagram of an improved Euler formula calculation model in a simulation analysis method for flashover tripping of a transmission line proposed by the present invention;

[0030] Figure 7 A schematic flow chart of a simulation analysis method for flashover tripping of a transmission line proposed by the present invention;

[0031] Figure 8 A simulation analysis method for transmission line flashover tripping proposed in the present invention obtains a sound field distribution model and a schematic diagram of the sound field time domain change process. DETAILED DESCRIPTION

[0032] The following will be combined with the accompanying drawings of the present invention to clearly and completely describe the simulation technology route in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] In recent years, with the widespread application of machine learning technology, the field of object detection has ushered in rapid development. However, in the field of transmission lines, especially in the detection of transmission line flashover tripping, the existing technology can usually only provide a preliminary judgment on the fault, and it is difficult to conduct in-depth comprehensive analysis. In response to this challenge, the present invention proposes an innovative simulation analysis method for transmission line flashover tripping. This method uses a multi-physics field finite element simulation model to accurately determine the acoustic field distribution map and time domain changes under different short-circuit conditions. Through these data, reliable data support can be provided for the distribution of air micro-vibration sensors and fault point location. The core principle of this method is to use the interaction of multiple physical fields such as electric field, thermal field and acoustic field to simulate the sound signal and vibration characteristics generated when the transmission line is short-circuited. Compared with traditional methods, this analysis method based on a multi-physics field finite element simulation model has higher precision and accuracy. It can quickly and accurately predict the occurrence of transmission line tripping flashover events and perform intelligent diagnosis. By analyzing the acoustic field distribution diagram and time domain changes, we can gain a deep understanding of the transmission line behavior under different short-circuit conditions, thereby timely discovering potential fault points and taking appropriate measures for repair and maintenance, greatly improving the stability and reliability of the transmission line. Figure 1 and Figure 7, the method specifically comprises the following steps:

[0035] S1, constructing a mathematical short-circuit model of a transmission line, wherein the mathematical short-circuit model of a transmission line is formed by constructing a relationship between impulse wave parameters (such as impulse voltage and impulse current) and short-circuit conditions, thereby obtaining current and voltage waveforms and fault duration of the transmission line under different short-circuit conditions and impulse wave parameters;

[0036] In step S1, a mathematical short-circuit model of the transmission line is established by conducting short-circuit circuit, arc heat release and thermal expansion acoustic theory analysis. The construction of this model involves multiple theoretical principles and analysis methods.

[0037] Specifically, first, a short-circuit circuit theory analysis was conducted. The essence of a transmission line fault is the insulation breakdown or short-circuit between conductors, which leads to abnormal increase in current. Considering the relationship between conductors with different potentials, short-circuit faults are divided into three-phase short circuits and two-phase short circuits, and possible causes are analyzed, such as the closing of the line with ground wire and the three-phase ground short circuit caused by the pole falling. These abnormal current and voltage fluctuations are one of the bases for establishing a mathematical model; second, an arc heat release theory analysis was conducted. In low-voltage distribution lines, arc short circuit faults are often caused by power supply short circuits, equipment aging, and protection equipment failure. The large amount of heat generated by arc heat release will cause the current to increase instantly and generate a large amount of electricity and heat in a short time. These heat release processes have an important impact on the stability and safety of the circuit, and are also one of the factors that need to be considered when establishing a mathematical model; third, a thermal expansion theory analysis was conducted. The large amount of heat generated by arc heat release will cause the temperature of the material to increase, thereby causing thermal expansion. Thermal expansion is closely related to the physical properties of the material, such as heat capacity and melting point. Energy conversion and material property changes during thermal expansion also have a significant impact on the fault behavior of transmission lines. When establishing a mathematical short-circuit model, this application introduces shock wave parameters, such as impulse voltage and impulse current. There is a complex relationship between these parameters and short-circuit conditions. For details, please refer to Figure 3 As shown in the figure, a mathematical short-circuit model of the transmission line is established by using short-circuit analysis, arc heat release and thermal expansion analysis. Shock wave parameters such as impulse voltage and impulse current are introduced into the model to analyze their relationship with short-circuit conditions. Short-circuit circuit analysis and thermal expansion analysis are performed. Under different short-circuit conditions, the resistance current of the circuit conductor is measured to increase. According to the formula Q = I 2Rt, and the relationship between arc heat release, heat increases, temperature rises, and thermal expansion occurs. The severe thermal shock caused by short-circuit circuits under different conditions that causes circuit tripping can be calculated. The relationship between current and thermal shock under short-circuit conditions can be used to explore the internal connection between short-circuit conditions and shock wave parameters, and find out the main condition factors affecting shock wave parameters. The arc heat release relationship is based on the arc heat release simulation model Mayr model, which can be deduced as: Formula 1 describes the rate of change of the instantaneous conductance g of the arc with time. When the arc is discharging, the change of conductance g is affected by the steady-state arc conductance G and the arc time constant T. T represents the time for the arc to decay to its initial conductance; Formula 2 describes the relationship between the arc time constant T and the peak current I of the arc and the arc length L. The arc time constant T represents the time for the arc to decay from the peak current to its initial conductance, and α is a constant; Formula 3 describes the relationship between the steady-state arc conductance G and the instantaneous current i of the arc. The arc conductance G is affected by the instantaneous current i of the arc and the static arc voltage drop V per unit length. After obtaining the simulation circuit current I, it is substituted into the above formula to calculate the arc time constant T. Next, the dynamic resistance is corrected in each time step, and the form of the time-varying resistance is controlled, and it is sent to the simulation circuit to solve the differential equation. In one or more embodiments, such as Figure 6 As shown, the improved Euler formula method is used to solve the differential equation. The specific form is as follows: According to the arc heat release model (Mayr model), the dynamic resistance is corrected by the Euler formula method in each time step, where G is the steady-state arc conductance, g(t) is the instantaneous conductance, T is the arc time constant, and v is the arc speed. In order to ensure the stability of the numerical calculation, the Taylor series expansion of the exponential function exp(x) is used: Taylor series expansion is used to calculate the approximate value of the exponential function to ensure the stability of the numerical calculation. The first two variable behaviors of the expansion are taken to obtain the stable behavior of the arc dynamic conductivity, which is the mathematical short-circuit model of the transmission line: According to the Taylor series expansion, the dynamic resistance is corrected, and the form of the time-varying resistance is controlled so that the differential equation can be stably solved in the simulation circuit. During the simulation process, key data, including current and voltage waveforms and fault duration information, were recorded. In addition, an experimental device was built near the transmission line to record the changes in short-circuit voltage, current waveforms, and shock wave parameters during the short-circuit process, as well as the behavioral response of the transmission line. The experimental data was then compared in detail with the simulation results to verify the accuracy of the simulation results. Finally, the simulation results were used to predict and analyze the flashover tripping of the transmission line.

[0038] Through the construction of mathematical models, we can deeply understand the impact of these parameters on the transmission line, and then predict key data such as current, voltage waveforms and fault duration under different short-circuit conditions and shock wave parameters. After the mathematical short-circuit model of the transmission line is constructed, this model is further used for simulation analysis. Different short-circuit conditions and shock wave parameters are adjusted, and their impact on the transmission line is observed. Key data, including current, voltage waveforms and fault duration, are recorded during the simulation process. The key to this step is to verify and analyze the accuracy and effectiveness of the mathematical model through simulation. By adjusting different short-circuit conditions and shock wave parameters, various possible situations are simulated, and the simulation results are recorded. These simulation results provide a basis for subsequent data analysis and fault diagnosis. In general, through step S1, a comprehensive mathematical model is established to describe the behavior of the transmission line. After the model is simulated and analyzed, the current, voltage waveforms and fault duration details of the transmission line under different short-circuit conditions and shock wave parameters are obtained.

[0039] S2, comprehensively analyze the current, voltage waveforms and fault duration of the transmission line under different short-circuit conditions and shock wave parameters obtained through the mathematical short-circuit model of the transmission line and the experimental data, so as to obtain the short-circuit conditions and shock wave size when the transmission line trips;

[0040] In step S2, the verification of the mathematical short-circuit model of the transmission line and the comprehensive analysis of the experimental data will be carried out. This step aims to determine the short-circuit conditions and shock wave size when the transmission line trips by comparing the predicted results of the mathematical model with the actual experimental data. Specifically, the mathematical model will be used to simulate the circuit response under different short-circuit conditions and shock wave parameters. The mathematical model is constructed based on short-circuit circuit theory, arc heat release mechanism and thermal expansion acoustic principle. By considering the influence of abnormal current increase, voltage fluctuation and heat release process on circuit stability, the fault behavior of the transmission line is deeply understood and its response under different conditions is predicted. Secondly, experimental data is collected to verify the accuracy of the mathematical model. In one or more embodiments, high-speed cameras and current and voltage sensor equipment are placed on site to record the changes of various parameters during the short-circuit process. The transmission line response in the real scene is obtained through experimental data, which provides basic data for subsequent comprehensive analysis. In the comprehensive analysis stage, the predicted results of the mathematical model are compared with the experimental data. By analyzing the difference between the two, in one or more embodiments, the actual short-circuit conditions and shock wave size when the transmission line trips can be determined through root mean square error (RMSE), mean absolute error (MAE), scatter plot, correlation analysis, and residual analysis. It is worth mentioning that the above methods are all conventional data analysis methods, so we will not go into details here. This process is an important step in verifying the reliability of the mathematical model, and it also provides a reliable reference for the operation and maintenance of the actual line. In the comprehensive analysis, external environmental factors and equipment status influencing factors will also be considered. By comprehensively considering the theoretical model and experimental data, the behavioral characteristics of the transmission line can be more comprehensively understood, providing accurate information and data support for fault diagnosis and line status monitoring.

[0041] S3, predicting the short-circuit condition of the transmission line through a simulation model under the short-circuit condition, thereby obtaining prediction data, wherein the prediction data is the tripping probability under different short-circuit conditions;

[0042] In step S3, the known short-circuit conditions and shock wave size when the transmission line trips are used to perform simulation to evaluate the possibility of transmission line tripping under different short-circuit conditions, that is, the tripping probability. Specifically, with the help of the previously established mathematical short-circuit model of the transmission line, various possible short-circuit conditions and shock wave parameters are set in the simulation software, such as short-circuit location, type (single-phase, two-phase or three-phase short circuit), impulse voltage and impulse current during short circuit. Simulation is performed to simulate a variety of possible short-circuit conditions by changing these parameter conditions, and observe the response of the transmission line under different conditions. During the simulation process, key data, including current, voltage waveform, and fault duration, are recorded for subsequent analysis. Then, the recorded simulation data is analyzed in detail to compare the behavior characteristics of the transmission line under different short-circuit conditions and gain an in-depth understanding of its fault behavior. Based on the analysis, the calculated tripping probability is used as the input data of the next step convolutional neural network, providing important feature information for the neural network. These probability values ​​reflect the expected degree of transmission line tripping events under different conditions, and are the key inputs for neural network fault diagnosis.

[0043] S4, obtaining acoustic signal data generated during a short circuit, and inputting the acoustic signal data and the predicted data into a convolutional neural network to obtain feature data, wherein the acoustic signal data is obtained by arranging an acoustic sensor near the transmission line;

[0044] Specifically, before collecting the acoustic signal data of the transmission line, it is necessary to arrange acoustic sensors near the transmission line. These acoustic sensors will be installed at a position close to the transmission line to ensure that the acoustic signals generated during the short circuit can be captured. The arrangement of the sensors will take into account the coverage, sensitivity and arrangement height factors of the sensors to ensure that the acoustic signals can be effectively captured. Once the acoustic sensors are arranged, they begin to collect the acoustic signal data generated during the short circuit. These acoustic signal data contain the sounds generated during arc discharge, usually including crackling sounds and arc vibration sounds. The collected acoustic signal data is used for fault diagnosis and status monitoring. The sound generated during arc discharge has specific frequency, amplitude and duration characteristics, which can reflect the fault condition of the transmission line. By analyzing the acoustic signal data, the location, type and severity of the fault can be determined, so that corresponding measures can be taken in time for repair and maintenance to ensure the safe operation of the transmission line.

[0045] The collected sound signal data is preprocessed so that it can be input into the convolutional neural network. Preprocessing includes normalization, dataset download and loading steps, please refer to Figure 4 As shown, it is worth mentioning that the following method is only one way to implement the process idea, including:

[0046] Step 1 Normalization: Use PyTorch to write a normalization indicator function to normalize the sound signal data to ensure that the data is within an acceptable range, which helps improve the training effect of the model.

[0047] Step 2 Dataset download: Download the corresponding sound signal dataset from the dataset MN I ST. When downloading the dataset, you need to set the storage path and specify whether it is a training set or a test set.

[0048] Step 3: Load the dataset: Instantiate a dataset object and use DataLoader to package the dataset for loading data.

[0049] After preprocessing, start building the CNN model. Figure 5 In this embodiment, a convolutional neural network model is constructed to extract features from sound signal data. This process includes setting up structures such as convolutional layers, pooling layers, and fully connected layers:

[0050] according to Figure 5 The convolutional neural network structure shown in the figure is constructed. The convolution layer (CONV) sets hyperparameters, including the number of filters, stride, and padding selection. Among them, the number of filters determines the number of convolution kernels, and each convolution kernel can capture different features in the input data. The stride determines the sliding step of the convolution kernel on the input data, which affects the size of the output. The padding method selects the 'Same' method, that is, zero padding is performed around the input data to keep the size unchanged after convolution. For example, for an input data set of (8,8), when performing a (3,3) filter convolution, zero padding will be performed around it, so that the convolution result is still (8,8) in size. The pooling layer is used to reduce the spatial dimension of the feature map to reduce the number of parameters and extract the main features. The description mentions that Maxpooling is used, and a 2x2 window is used for operation with a stride of 2. This means that in each 2x2 window, the maximum value is taken as the output of the window, and then moved by 2 strides. This can effectively reduce the size of the feature map while retaining the most important features. The activation function ReLU is applied after the output of the convolutional layer. The ReLU function sets negative values ​​to zero and retains positive values, introducing nonlinear characteristics that enable the network to learn more complex functions. After passing through the convolutional layer and the pooling layer, the resulting feature map is flattened through a view operation, converted into a one-dimensional tensor, and then input into the fully connected layer. The fully connected layer is used to map high-dimensional features to the output dimension so that it can be connected to the next input. The input and output of the fully connected layer are both two-dimensional tensors, that is, a one-dimensional vector.

[0051] The prediction data obtained in step S3, i.e., the tripping probability under different short-circuit conditions, also needs to be input into the convolutional neural network. These prediction data will be input into the neural network together with the acoustic signal data to enhance the prediction ability of the model and the fault diagnosis effect.

[0052] Through the processing of convolutional neural networks, feature information is extracted from the sound signal data and predicted data. These feature data are abstract representations of the input data and contain key information in the input data. These features are the frequency, amplitude and other features in the sound signal, and the trip probability in the predicted data. Thus, features are extracted from the sound signal data generated during a short circuit, and combined with the predicted data, feature data representing the input data is obtained through a convolutional neural network. These feature data can be used for the subsequent processing of the input end of the multi-physics field finite element simulation model, and for the analysis of the sound field distribution model and the time domain changes of the sound field.

[0053] S5, inputting the characteristic data into a multi-physics field finite element simulation model, thereby obtaining a sound field distribution model and a time domain change of the sound field in the vicinity of the transmission line short circuit;

[0054] In step S5, the characteristic data is used to input the situation near the short circuit of the transmission line into the multi-physics finite element simulation model. In one or more embodiments, the multi-physics finite element simulation model includes three physical fields: electric field, thermal field and acoustic field, and a coupling relationship is established between them. Through the simulation of multiple physical fields, the sound field distribution model and the time domain change of the sound field in the vicinity of the short circuit of the transmission line can be obtained. Specifically, the model of each physical field is established by finite element software, and the corresponding boundary conditions and coupling conditions are set. Then the model is parameter optimized and meshed to obtain more accurate results. Finally, the distribution model and time domain change of the sound field are obtained, which provides important data support for further analysis and research.

[0055] Specifically, refer to Figure 2 , Figure 2 The invention discloses a multi-physics simulation technology roadmap for flash explosion phenomenon, which includes a multi-physics model and a simulation result processing model. The input end of the multi-physics finite element simulation model of flash explosion phenomenon is connected to the output end of a convolutional neural network.

[0056] The multi-physics field model uses finite element software to establish three physical fields: electric field, thermal field and acoustic field, and sets physical field coupling conditions and boundary conditions. The physical field coupling conditions involve the interaction between different physical fields. The finite element model uses displacement constraint coupling to constrain the displacements of adjacent units together to ensure the continuity and stability of the multi-physics field model. In addition, electric-thermal coupling is established and electric-thermal coupling conditions are set. According to the Morgan formula, the current carrying capacity of the short-circuit circuit conductor is known, and the surface temperature of the short-circuit circuit conductor can be calculated. The current carrying capacity of the conductor can be calculated if the surface temperature of the conductor is known. The formula is simplified as follows:

[0057] Where: W T =0.57πλγ(TT a ) 0.435 , is the convection heat dissipation rate per unit length of conductor (w / m);

[0058] W Z =α i J i D, is the solar absorption rate per unit length of conductor (w / m);

[0059] W R =πεSD[(T+273) 4 -(T a +273) 4 ], is the radiation heat dissipation rate per unit length of wire;

[0060] Where:

[0061] Re——Reynolds coefficient

[0062] T——wire temperature

[0063] Ta——ambient temperature

[0064] D——Outer diameter of the conductor

[0065] V——wind speed perpendicular to the conductor

[0066] γ——The kinematic viscosity of the air on the conductor surface,

[0067] α i ——The heat absorption coefficient of the conductor is equal to the radiation coefficient

[0068] J i ——The intensity of sunlight on the conductor

[0069] ε——Radiance of conductor surface, bright new line is 0.23-0.46

[0070] S——Stefan-Boltzmann coefficient, 5.67x10-8

[0071] Morgan's formula is used to calculate the current carrying capacity I of a short-circuit conductor, which takes into account the effects of convective heat dissipation rate, solar absorption rate and radiation heat dissipation rate on the conductor's current carrying capacity. These parameters describe the heat dissipation, heat absorption and radiation heat dissipation of the conductor, respectively, which in turn affect the surface temperature and current carrying capacity of the conductor.

[0072] A thermal-acoustic coupling model was established, and the sound pressure boundary conditions were set. Then, the finite element software was used to iterate between the two modules representing the thermal field and the acoustic field when the transmission line was short-circuited to simulate the mutual influence between heat and sound. Specifically, the third type of boundary conditions of Rob in were adopted in this embodiment. This boundary condition allows the determination of numerical and arbitrary function conditions to better adapt to the simulation model. Next, a suitable grid division was established based on the model theory of each physical field. Because good grid division can improve the accuracy and stability of the simulation model. The calculation results were then parameter optimized, and finally the sound field distribution diagram near the short-circuit area and the time domain change results of the sound field were obtained. The derivation process is as follows. When deriving the sound wave signal equation, the propagation of the sound wave is considered. The sound wave signal equation can be expressed as:

[0073] Where u is the sound wave signal, t is time, x is the spatial coordinate, v is the speed of sound, and f is the external excitation. Multiply the sound wave signal equation by a basis function And integrate over the space: The basis functions described here is a piecewise linear function, often called a linear interpolation function or a hat function. It is linear within each segment and continuous between adjacent segments. Specifically, when x i-1 <x<x i When The value of This means that in the interval [x i-1 ,x i ], the basis function value is in x i-1 is 1 at x i is 0 at the end and changes linearly between the two endpoints; when x i <x<x i+1 , basis function The value of This means that in the interval [x i ,x i+1 ], the basis function value is in x i is 0 at x i+1 1 at the end point and varies linearly between the two endpoints; in other cases, x<x i-1 or x>x i+1When The value is 0, which means that in the interval [x i-1 ,x i ] and [x i ,x i+1 ], the basis function takes the value of 0, which does not affect the expression of the solution. Next, the spatial terms in the acoustic wave equation are processed by the method of integration by parts. Specifically, the basis function Discretize the spatial term. Apply the method of integration by parts to process the second derivative of the spatial term and obtain: Here, for the second derivative The first term in the result is the contribution of the discretized spatial term, and the second term is the contribution of the boundary term. Assuming that the boundary u(x1)=u(x2)=0, use the basis function to interpolate and approximate it, which represents the fixed boundary condition of the sound wave at the boundary. Next, we use the basis function The displacement on the boundary is interpolated and approximated, so that u(x) can be discretized in the entire area. Therefore, u(x) is represented by a linear combination of basis functions: Then the spatial terms and external excitation terms f in the acoustic wave equation are discretized in the region and combined with the boundary conditions. The following expression is obtained by interpolating u(x) to the boundary: The first and second terms represent the contribution of the discretized spatial term and the external excitation term in the region, respectively. The third term is the contribution of the boundary condition u(x1) = u(x2) = 0. Next, the second-order difference format (implicit difference format) is used to discretize the time term of the acoustic wave equation. The time term is discretized using the second-order difference format. In this process, the time derivative Replaced with the second-order time difference to better approximate the rate of change of the time derivative. The implicit difference format is used here because it is more conducive to numerical stability. Then the time term in the acoustic wave equation is discretized and combined with the spatial term, external excitation term and boundary conditions. By discretizing the time term using the second-order difference format, the following expression is obtained: The first term represents the discretization of the time term by the second-order difference format. The second term represents the coupling effect between the time term and the space term in the acoustic wave equation, where represents the displacement at time step n+1, is the rate of change over time. Further simplification into the form of matrix multiplication is as follows: (M 2 +v 2 Δt 2 ) n+1 =M(2U n -U n-1 )+Δt 2F, where M is called the mass matrix, S is called the stiffness matrix, and the F vector represents the source term. The acoustic wave equation can be simplified to a matrix equation by expressing it in a matrix, where (M 2 +v 2 Δt 2 ) is a matrix that represents the combination of time and space terms, M is the mass matrix, and U n+1 is the unknown displacement vector of the next time step, U n and U n-1 is the known displacement vector of the current and previous time step, and F is the external excitation term vector. Finally, after a simple integration, the specific values ​​in the two matrices can be obtained, namely:

[0074]

[0075] Among them, the first integral formula describes the spatial term in the sound field equation, that is, the spatial distribution model of the sound field. This formula calculates the numerical representation of the sound field at different positions by integrating the basis function, which includes the cross terms and local terms of the sound field at different positions. The spatial distribution of the sound field in the area near the transmission line short circuit can be obtained by calculation of this formula. The second integral formula describes the time term in the sound field equation, that is, the evolution of the sound field over time. This formula calculates the numerical representation of the sound field at different time steps by integrating the basis function, which includes the differential terms of the sound field at the current moment and the previous moment. The time domain changes of the sound field evolving over time can be obtained by calculation of this formula. Combining these two integral formulas, the spatial distribution model and time domain changes of the sound field can be expressed in numerical form, so as to fully understand the sound field behavior in the area near the transmission line short circuit. These numerical representations provide specific numerical solutions to the sound field, which provide an important basis for analyzing the behavior of the sound field and evaluating acoustic performance. In general, step S5 includes the following steps:

[0076] S51, establishing a multi-physics field finite element simulation model, wherein the multi-physics field finite element simulation model uses finite element software to establish three physical fields: electric field, thermal field and acoustic field;

[0077] S52, performing electrothermal coupling and thermoacoustic coupling settings on the multi-physics field finite element simulation model;

[0078] S53, based on the multi-physics field model and coupling setting, deriving the acoustic wave signal model and discretizing it;

[0079] S54, obtaining the spatial distribution model and time domain changes of the sound field through the sound wave signal model.

[0080] S6, visualize or output the acoustic field distribution near the short circuit of the transmission line and the time domain change results, so as to perform status monitoring and fault diagnosis of the transmission line.

[0081] Specifically, in step S6, the obtained sound field distribution near the short circuit of the transmission line and the time domain change results need to be visualized or output. This includes the spatial visualization of the sound field distribution, and the generation of time domain curves or dynamic graphs of the sound field changing over time. Specific visualization methods may include two-dimensional or three-dimensional charts, heat maps, animations, etc., to intuitively display the distribution and evolution of the sound field. The results of visualization or output provide a theoretical and data basis for the design of air micro-vibration sensors. By observing the distribution and time domain changes of the sound field, the appropriate sensor layout position, as well as the appropriate sampling frequency and parameter settings are determined. Such information is of great significance for designing and optimizing the algorithms and parameters of air micro-vibration sensors. The ultimate goal is to use air micro-vibration sensors to monitor the status of the transmission line and diagnose faults. By monitoring the changes in the sound field, abnormal conditions in the line can be discovered in time, and real-time monitoring of the status of the transmission line and timely diagnosis of faults can be achieved.

[0082] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A simulation analysis method for transmission line flashover tripping, characterized in that: The following steps are involved: S1, constructing a mathematical short-circuit model of a transmission line, wherein the mathematical short-circuit model of a transmission line is formed by constructing a relationship between a shock wave parameter and a short-circuit condition, thereby obtaining current and voltage waveforms and fault duration of the transmission line under different short-circuit conditions and shock wave parameters; S2, comprehensively analyze the current, voltage waveforms and fault duration of the transmission line under different short-circuit conditions and shock wave parameters obtained through the mathematical short-circuit model of the transmission line and the experimental data, so as to obtain the short-circuit conditions and shock wave size when the transmission line trips; S3, predicting the short-circuit condition of the transmission line through a simulation model under the short-circuit condition, thereby obtaining prediction data, wherein the prediction data is the tripping probability under different short-circuit conditions; S4, obtaining acoustic signal data generated during a short circuit, and inputting the acoustic signal data and the predicted data into a convolutional neural network to obtain feature data, wherein the acoustic signal data is obtained by arranging an acoustic sensor near the transmission line; S5, inputting the characteristic data into a multi-physics field finite element simulation model, thereby obtaining a sound field distribution model and a time domain change of the sound field in the vicinity of the transmission line short circuit; S6, visualizing or outputting the acoustic field distribution and time domain variation results near the short circuit of the transmission line, so as to perform status monitoring and fault diagnosis of the transmission line; The impulse wave parameters in step S1 include impulse voltage and impulse current; the specific calculation formula of the transmission line mathematical short-circuit model is: Where g(t) represents the instantaneous conductance of the arc, g(t+Vt) represents the value of the instantaneous conductance of the arc after a period of time step vT, [Gg(t)] represents the difference between the steady-state conductance G of the arc and the instantaneous conductance g(t), The correction term represents the decay trend of the arc dynamic conductivity over time.

2. A simulation analysis method for transmission line flashover tripping according to claim 1, characterized in that: The experimental data in step S2 is obtained by placing a high-speed camera and a current and voltage sensor near the transmission line to record the short-circuit voltage, current waveform, changes in shock wave parameters and transmission line behavior response data during the short-circuit process.

3. The simulation analysis method for transmission line flashover tripping according to claim 1 is characterized in that: The step S4 pre-processes the acoustic signal data and the prediction data before inputting them into the convolutional neural network, and the pre-processing includes normalization, data set downloading and loading steps.

4. A simulation analysis method for flashover tripping of a transmission line according to any one of claims 1 to 3, characterized in that: The step S5 comprises the following steps: S51, establishing a multi-physics field finite element simulation model, wherein the multi-physics field finite element simulation model uses finite element software to establish three physical fields: electric field, thermal field and acoustic field; S52, performing electrothermal coupling and thermoacoustic coupling settings on the multi-physics field finite element simulation model; S53, based on the multi-physics field model and coupling setting, deriving the acoustic wave signal model and discretizing it; S54, obtaining the spatial distribution model and time domain changes of the sound field through the sound wave signal model after discrete processing.

5. A simulation analysis method for transmission line flashover tripping according to claim 4, characterized in that: The specific calculation formula of the acoustic wave signal model is: Among them, u is the sound wave signal, t is time, x is the spatial coordinate, v is the speed of sound, and f is the external excitation.

6. A simulation analysis method for transmission line flashover tripping according to claim 4, characterized in that: The specific calculation formula of the acoustic wave signal model after discrete processing is (M 2 +v 2 Δt 2 ) n+1 =M(2U n -U n-1 )+Δt 2 F, where M is the mass matrix, v is the speed of sound, Δt is the time step, and U n+1 is the unknown displacement vector of the next time step, U n and U n-1 are the known displacement vectors of the current and previous time steps respectively, and the F vector represents the source term.

Images