Power transmission line hidden danger identification method and system

Abstract

The invention discloses a power transmission line hidden danger identification method and system, and relates to the technical field of power transmission line hidden danger identification, and the method comprises the steps: collecting traveling wave current and power frequency voltage data in a simulation test period of a specific voltage grade, and recording a discharge distance corresponding to the traveling wave current; performing time domain and frequency domain feature extraction on the traveling wave current waveform, constructing a data matrix and analyzing hidden danger features; and carrying out time sequence analysis on hidden danger development, and identifying early indexes of hidden danger development. According to the method, the state transition and future trend prediction capability of the hidden Markov model are combined, so that the development trend of the hidden danger of the power transmission line can be predicted in time, preventive measures can be taken in advance, and the safety and stability of a power transmission system are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission line hidden dangers, and in particular to a power transmission line hidden danger identification method and system. Background Art

[0002] Transmission lines cross mountains and rivers and are extremely susceptible to tripping accidents caused by the surrounding natural environmental factors such as lightning, dirt, plants and animals, and wind. Every trip will have an impact on the power grid and cause damage to facilities such as wires and insulators. Before tripping, there is usually a period of abnormal discharge state. This unsafe operating state is defined as a hidden danger. Scholars have done a lot of research to identify hidden dangers in a timely manner and deal with them, basically focusing on the field of image recognition.

[0003] For the status monitoring of power transmission lines, both at home and abroad, drone inspections are basically adopted. Drones are equipped with cameras, lidars, infrared equipment and other tools to inspect and collect data on the transmission lines, and obtain high-precision and high-density three-dimensional spatial information of the terrain and objects near the transmission line corridor. The collected raw data is transmitted to the workstation of the ground terminal through communication equipment. The data processing personnel use professional software to open the collected data, process the data, and analyze the defects and hidden dangers in the transmission line. This intelligent line detection technology saves a lot of manpower, but these technologies often have the problems of single applicability, high cost, and low recognition accuracy. In response to these shortcomings, we have invented a method and system for identifying hidden dangers in transmission lines. It installs a traveling wave monitoring device on the overhead line, extracts the time domain and frequency domain parameter characteristics of the detected current traveling wave pulse discharge waveform, and can identify the type of hidden danger without image recognition. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above problems or problems existing in the prior art, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that the prior art lacks effective state transfer and future trend prediction capabilities when processing time series data, resulting in the inability to take preventive measures in a timely manner.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for identifying hidden dangers of a power transmission line, which comprises collecting traveling wave current and power frequency voltage data during a simulation test of a specific voltage level, and recording the discharge distance corresponding to the traveling wave current;

[0008] Extract the time domain and frequency domain features of the traveling current waveform, construct a data matrix and analyze the hidden danger characteristics;

[0009] Conduct time series analysis on hidden danger development and identify early indicators of hidden danger development.

[0010] As a preferred solution of the power transmission line hidden danger identification method of the present invention, wherein: the recording of the discharge distance corresponding to the traveling wave current includes determining the critical discharge distance of different hidden danger types and the time when they are located in combination with an ultraviolet imager.

[0011] As a preferred solution of the method for identifying hidden dangers of power transmission lines of the present invention, the extraction of time domain and frequency domain features of the traveling current waveform includes:

[0012] Pre-process the collected traveling current waveform to remove noise and interference;

[0013] Apply wavelet transform technology to extract the subtle change characteristics of the traveling current waveform in the time domain;

[0014] The Fourier transform method is used to obtain the distribution characteristics of the traveling wave current waveform in the frequency domain;

[0015] The extracted time domain and frequency domain features are normalized;

[0016] The constructing of the data matrix comprises:

[0017] The extracted time domain features and frequency domain features are fused to form a comprehensive feature vector;

[0018] Construct a data matrix containing multiple eigenvectors.

[0019] As a preferred solution of the method for identifying hidden dangers of power transmission lines of the present invention, the analysis of hidden danger characteristics includes:

[0020] The data matrix is ​​processed for dimensionality reduction to reduce the data dimension and extract the main components, which can be expressed as:

[0021]

[0022] Among them, Z is the feature matrix after dimensionality reduction, W is the feature matrix, X is the original data matrix, α and β are adjustment parameters, σ i is the standard deviation of the data sample, g(x) is a weighted kernel function used to enhance feature extraction;

[0023] Furthermore, the kernel function g(x) is expressed as:

[0024]

[0025] Where x is the input data point; x j is the center point of the kernel function; w j is the kernel function weight; σ is the width parameter of the kernel function; M is the number of kernel function center points;

[0026] Classify the feature vector after dimensionality reduction and identify the key features of the hidden danger type, which can be expressed as:

[0027]

[0028]

[0029] Among them, S w is the intra-class scatter matrix, S b is the inter-class scatter matrix, x i is the data sample, C is the number of categories, N c is the number of samples of class c, μ c is the class mean, γ and δ are adjustment parameters, σ represents the overall standard deviation, f(x i ) nonlinear mapping function, h(μ c ,μ) is the nonlinear mapping function, J is the matrix transpose;

[0030] Furthermore, f(x i ) and h(μ c ,μ) are respectively expressed as:

[0031] f(x i )=tanh(ax i +b)

[0032]

[0033] Among them, a and b are the parameters of the mapping function; μ c is the class mean; μ is the overall mean; δ is the smoothing parameter to prevent the denominator from being zero;

[0034] According to the analysis results, the main components and key features are extracted, and the characteristic models of different hidden danger types are established, which are expressed as:

[0035]

[0036] Among them, M is the feature model, λ i is the weight of the PCA principal component, z i is the feature after dimensionality reduction, η j is the weight of the LDA key feature, s j Key features identified by LDA.

[0037] As a preferred solution of the transmission line hidden danger identification method described in the present invention, the time series analysis of hidden danger development and identification of early indicators of hidden danger development include modeling the time series data of traveling wave current through a hidden Markov model to predict the trend and timing of hidden danger development.

[0038] As a preferred solution of the method for identifying hidden dangers of power transmission lines of the present invention, the modeling includes:

[0039] Arrange the traveling wave current data in chronological order to form a time series data set O = {o 1 ,o 2 ,…,o T},o t represents the observed value at time t, and T represents the total time length;

[0040] Standardize the data, calculate the mean and standard deviation of each observation, and obtain the standardized data ot′;

[0041] Define the state space and observation sequence respectively, expressed as:

[0042] S={S 1 ,S 2 ,…,S N}

[0043] O={o 1 ′,o 2 ′,…,o T ′}

[0044] Among them, S i represents the i-th state, there are N possible states in total; ot′ is the standardized observation value;

[0045] Initialize the state probability vector, expressed as:

[0046]

[0047] Among them, π i Indicates that the system is in state S at time t = 1 i The probability of i is the basic initial state probability parameter, κ is the adjustment parameter, and λ is the frequency;

[0048] Define the state transition probability matrix A = {a ij}, where a ij Indicates that from state S i Transfer to state S j The probability is expressed as:

[0049] Among them, α ijis the basic transition probability parameter, β is the adjustment parameter, ω is the frequency, and φ is the phase;

[0050] Define the observation probability matrix B = {b j (o t ′)}, where b j (o t ′) indicates that in state S j Observed below o t The probability of ′ is expressed as:

[0051]

[0052] Among them, μ j For state S j The observed mean, σ j For state S j The observed standard deviation, γ is the adjustment parameter, θ is the frequency, and δ is the phase;

[0053] Perform model training and parameter estimation, which includes,

[0054] Calculate the probability of the observed sequence using the forward algorithm;

[0055] Use the Viterbi algorithm to find the most likely sequence of states;

[0056] According to the results of the forward and Viterbi algorithms, the expectation-maximization algorithm is used to iteratively update the parameters of π, A, and B;

[0057] Predict the trend and timing of hidden danger development, including:

[0058] Given the current observation sequence O = {o 1 ′,o 2 ′,…,o T ′}, the observation probability at future moments under different states is calculated by the forward algorithm, which is expressed as:

[0059]

[0060] Predict the trend and time series of hidden danger development, and analyze the periodicity and nonlinear characteristics of hidden danger changes based on the time dependence term γcos(θt+δ) in the observation probability matrix B.

[0061] As a preferred solution of the power transmission line hidden danger identification method of the present invention, wherein: the probability of using the forward algorithm to calculate the observation sequence includes:

[0062] Initialize the forward variable α1(i):

[0063] α 1 (i) = π i b i (o1 ′)

[0064] Recursively calculate the forward variable αt(i):

[0065]

[0066] The method of using the Viterbi algorithm to find the most likely state sequence includes:

[0067] Initialize the Viterbi variable δ1(i):

[0068] δ 1 (i) = π i b i (o 1 ′)

[0069] Recursively calculate the Viterbi variable δt(i):

[0070] δ t (i) = max 1≤j≤N [δ t-1 (j)A ji ]B i (o t ′).

[0071] Another object of the present invention is to provide a system for identifying hidden dangers in transmission lines, which can collect traveling wave current and industrial frequency voltage data during simulation tests of specific voltage levels, record the discharge distance corresponding to the traveling wave current; extract time domain and frequency domain features of the traveling wave current waveform, construct a data matrix and analyze hidden danger characteristics; perform time series analysis on the development of hidden dangers, identify early indicators of hidden danger development, and solve the current problem of untimely identification of hidden dangers in transmission lines.

[0072] As a preferred solution of the system for identifying hidden dangers of power transmission lines of the present invention, it includes:

[0073] Acquisition module, extraction and analysis module, analysis module;

[0074] The acquisition module is used to collect the traveling wave current and power frequency voltage data during the simulation test of a specific voltage level, and record the discharge distance corresponding to the traveling wave current;

[0075] The extraction and analysis module is used to extract the time domain and frequency domain features of the traveling current waveform, construct a data matrix and analyze the hidden danger features;

[0076] The analysis module is used to perform time series analysis on the development of hidden dangers and identify early indicators of hidden danger development.

[0077] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a method for identifying hidden dangers of a power transmission line.

[0078] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for identifying hidden dangers of a power transmission line.

[0079] Beneficial effects of the present invention: The present invention achieves comprehensive coverage and accurate identification of hidden dangers in transmission lines by monitoring the multi-environmental conditions of traveling wave current and power frequency voltage data. Advanced signal processing algorithms such as wavelet transform and Fourier transform are used to extract time domain and frequency domain features, fully explore and utilize traveling wave current data, and improve the accuracy of hidden danger identification. The hidden Markov model is used to model time series data, and through the analysis of state transition and observation probability, the early development of hidden dangers can be identified and warned, effectively preventing further deterioration of hidden dangers. Combined with the state transition and future trend prediction capabilities of the hidden Markov model, the development trend of hidden dangers in transmission lines can be predicted in a timely manner, helping to take preventive measures in advance to ensure the safety and stability of the transmission system. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0081] Figure 1 This is the schematic diagram of the tree hazard experiment layout for the transmission line hazard identification method.

[0082] Figure 2 The on-site layout diagram of tree hazards for the transmission line hazard identification method.

[0083] Figure 3 This is the schematic diagram of the bird nest hazard experiment layout for the transmission line hazard identification method.

[0084] Figure 4 This is the on-site layout diagram of the floating object hazards for the transmission line hazard identification method.

[0085] Figure 5 Glass insulator layout diagram for transmission line hidden danger identification method.

[0086] Figure 6 Ceramic insulator layout diagram for transmission line hidden danger identification method.

[0087] Figure 7The figure is the overall flow chart of the hidden danger identification method for transmission lines.

[0088] Figure 8 This is a schematic diagram of the overall computer equipment. DETAILED DESCRIPTION

[0089] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0090] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0091] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0092] Example 1

[0093] Reference Figures 1 to 8 , is an embodiment of the present invention, and provides a method for identifying hidden dangers of a power transmission line, comprising:

[0094] S1: Collect the traveling wave current and power frequency voltage data during the simulation test of a specific voltage level, and record the discharge distance corresponding to the traveling wave current.

[0095] Furthermore, recording the discharge distance corresponding to the traveling wave current includes determining the critical discharge distances and the time points of different hidden danger types in combination with an ultraviolet imager.

[0096] Furthermore, the simulation experiment includes:

[0097] Taking the potential discharge of trees as an example, two types of trees (rubber tree and eucalyptus) were selected to obtain the discharge characteristics under different voltage levels (35kV, 110kV, 220kV) and different clearance distances under experimental environment. The experimental layout schematic diagram of trees and the field experimental platform are shown in the figure. Figure 1 As shown. The shifting coil voltage regulator, power frequency test transformer, protection resistor, AC capacitor voltage divider are sequentially connected according to the test layout principle as shown in Figure 1, connect the high-voltage wires in series and in parallel, and connect the oscilloscope probe to the voltage divider for voltage numerical observation. Hang a composite insulator on the overhead crane hook, and hang a wire on the other end of the hardware. The simulated wire is a homogeneous steel pipe with an outer diameter of 3.1cm and a length of 3m. A 3m*3m aluminum sheet is placed directly below to simulate the earth. Potted plants are placed directly below the steel pipe to simulate trees. The overall on-site layout is as follows Figure 2 , taking 35kV as an example, since the voltage is single-phase, it is added to 20kV, and the discharge distance is evenly changed, for example, 17cm-15cm-13cm-11cm, and the discharge situation is recorded.

[0098] Furthermore, the flashover steps of insulators are different. Glass insulators and ceramic insulators are selected. According to the artificial pollution test standard GB / T 4585-2004 for high-voltage insulators used in AC systems, the two types of insulators are mixed and smeared with contamination liquid. Here, one insulator is selected due to weight restrictions, and then the following steps are performed:

[0099] ① Check the circuit: Complete the test environment construction according to the test layout principle, and check and ensure that the high-voltage test circuit wiring is correct.

[0100] ② Sample wetting: Use a hand-operated sprayer to wet the surface of the dirty insulator, and use an industrial humidifier to simulate the mist. Figure 5 and Figure 6 .

[0101] ③ Voltage boost: operate the control console and slowly increase the voltage to the rated value according to the uniform voltage boost method. This test selects a piece of contaminated insulator as the test object. Under certain conditions, it simulates the discharge of ceramic and glass insulators at different voltages and different degrees of contamination, so as to obtain the discharge of contaminated insulators in the actual environment. This test will increase the voltage to 10kV, 15kV, 20kV, 25kV, 30kV, and 35kV respectively.

[0102] ④ Recording: After uniformly pressurizing to a certain voltage level, stay for 10s to 15s to ensure that the monitoring device can continuously record the discharge data at this voltage level. At the same time, use an ultraviolet imager, a visible light camera and an ultrasonic monitor to capture the actual phenomenon of contamination discharge.

[0103] ⑤ According to the test process of steps ① to ④, replace the ceramic and glass insulators with different degrees of contamination, and obtain the discharge data and actual phenomena of different types of insulators and different voltage levels.

[0104] S2: Extract the time domain and frequency domain features of the traveling wave current waveform, construct a data matrix and analyze the hidden danger characteristics.

[0105] Furthermore, the time domain and frequency domain feature extraction of the traveling wave current waveform includes:

[0106] Pre-process the collected traveling current waveform to remove noise and interference;

[0107] Apply wavelet transform technology to extract the subtle change characteristics of the traveling current waveform in the time domain;

[0108] The Fourier transform method is used to obtain the distribution characteristics of the traveling wave current waveform in the frequency domain;

[0109] The extracted time domain and frequency domain features are normalized;

[0110] The constructing of the data matrix comprises:

[0111] The extracted time domain features and frequency domain features are fused to form a comprehensive feature vector;

[0112] Construct a data matrix containing multiple eigenvectors.

[0113] Furthermore, the analysis of hidden danger characteristics includes:

[0114] The data matrix is ​​processed for dimensionality reduction to reduce the data dimension and extract the main components, which can be expressed as:

[0115]

[0116] Among them, Z is the feature matrix after dimensionality reduction, W is the feature matrix, X is the original data matrix, α and β are adjustment parameters, σ i is the standard deviation of the data sample, g(x) is a weighted kernel function used to enhance feature extraction;

[0117] Furthermore, the kernel function g(x) is expressed as:

[0118]

[0119] Where x is the input data point; x j is the center point of the kernel function; w j is the kernel function weight; σ is the width parameter of the kernel function; M is the number of kernel function center points;

[0120] Classify the feature vector after dimensionality reduction and identify the key features of the hidden danger type, which can be expressed as:

[0121]

[0122] Among them, S w is the intra-class scatter matrix, S b is the inter-class scatter matrix, x i is the data sample, C is the number of categories, N c is the number of samples of class c, μ cis the class mean, γ and δ are adjustment parameters, σ represents the overall standard deviation, f(x i ) nonlinear mapping function, h(μ c ,μ) is the nonlinear mapping function, J is the matrix transpose;

[0123] It should be noted that the range of linear discriminant analysis is the classification boundary of the feature vector, which represents the scatter between different categories. The range of the range depends on the specific calculation results of the intra-class scatter matrix and the inter-class scatter matrix, and is mainly used to optimize the classification effect and identify key features.

[0124] Furthermore, f(x i ) and h(μ c ,μ) are respectively expressed as:

[0125] f(x i )=tanh(ax i +b)

[0126]

[0127] Among them, a and b are the parameters of the mapping function; μ c is the class mean; μ is the overall mean; δ is the smoothing parameter to prevent the denominator from being zero;

[0128] According to the analysis results, the main components and key features are extracted, and the characteristic models of different hidden danger types are established, which are expressed as:

[0129]

[0130] Among them, M is the feature model, λ i is the weight of the PCA principal component, z i is the feature after dimensionality reduction, η j is the weight of the LDA key feature, s j Key features identified by LDA.

[0131] It should be noted that by establishing the feature model and analyzing the output results, different types of transmission line hazards can be accurately identified and classified. Each type of hazard has its own unique characteristics, and by combining PCA and LDA, these characteristics can be extracted and effectively classified.

[0132] S3: Conduct time series analysis on the development of hidden dangers and identify early indicators of hidden danger development.

[0133] Furthermore, the time series analysis of hidden danger development and identification of early indicators of hidden danger development include modeling the time series data of traveling wave current through a hidden Markov model to predict the trend and timing of hidden danger development.

[0134] Furthermore, the modeling includes:

[0135] Arrange the traveling wave current data in chronological order to form a time series data set O = {o 1 ,o 2 ,…,o T},o t represents the observed value at time t, and T represents the total time length;

[0136] Standardize the data, calculate the mean and standard deviation of each observation, and obtain the standardized data ot′;

[0137] Define the state space and observation sequence respectively, expressed as:

[0138] S={S 1 ,S 2 ,…,S N}

[0139] O={o 1 ′,o 2 ′,…,o T ′}

[0140] Among them, S i represents the i-th state, there are N possible states in total; ot′ is the standardized observation value;

[0141] Initialize the state probability vector, expressed as:

[0142]

[0143] Among them, π i Indicates that the system is in state S at time t = 1 i The probability of i is the basic initial state probability parameter, κ is the adjustment parameter, and λ is the frequency;

[0144] Define the state transition probability matrix A = {a ij}, where a ij Indicates that from state S i Transfer to state S j The probability is expressed as:

[0145] Among them, α ij is the basic transition probability parameter, β is the adjustment parameter, ω is the frequency, and φ is the phase;

[0146] Define the observation probability matrix B = {b j (o t ′)}, where b j (o t ′) indicates that in state Sj Observed below o t The probability of ′ is expressed as:

[0147]

[0148] Among them, μ j For state S j The observed mean, σ j For state S j The observed standard deviation, γ is the adjustment parameter, θ is the frequency, and δ is the phase;

[0149] Perform model training and parameter estimation, which includes,

[0150] Calculate the probability of the observed sequence using the forward algorithm;

[0151] Use the Viterbi algorithm to find the most likely sequence of states;

[0152] According to the results of the forward and Viterbi algorithms, the expectation-maximization algorithm (EM algorithm) is used to iteratively update the parameters of π, A, and B;

[0153] Predict the trend and timing of hidden danger development, including:

[0154] Given the current observation sequence O = {o 1 ′,o 2 ′,…,o T ′}, the observation probability at future moments under different states is calculated by the forward algorithm, which is expressed as:

[0155]

[0156] Predict the trend and time series of hidden danger development, and analyze the periodicity and nonlinear characteristics of hidden danger changes based on the time dependence term γcos(θt+δ) in the observation probability matrix B.

[0157] It should be noted that the method of using the forward algorithm to calculate the probability of the observation sequence includes:

[0158] Initialize the forward variable α1(i):

[0159] α 1 (i) = π i b i (o 1 ′)

[0160] Recursively calculate the forward variable αt(i):

[0161]

[0162] The method of using the Viterbi algorithm to find the most likely state sequence includes:

[0163] Initialize the Viterbi variable δ1(i):

[0164] δ 1 (i) = π i b i (o′ 1 )

[0165] Recursively calculate the Viterbi variable δt(i):

[0166] δ t (i) = max 1≤j≤N [δ t-1 (j)A ji ]B i (o t ′).

[0168] Furthermore, the Hidden Markov Model models the time series data and outputs a variety of results, such as the most likely state sequence, state transition probability, observation probability, initial state probability, and future state prediction. The following are methods and judgment processes for determining the development of several hidden dangers:

[0169] 1. The most likely state sequence

[0170] Output: a set of state sequences {q 1 ,q 2 ,…,q T}, where q t represents the most likely state at time t.

[0171] Determination method:

[0172] Normal state (S1): If most of the time points q t S1 means it is in normal state most of the time.

[0173] Potential hidden danger state (S2): If a certain period of time q t The transition from S1 to S2 and lasting for a period of time indicates that the system has entered a potential risk state.

[0174] Obvious hidden danger state (S3): If qt is S3 at most time points or qt changes rapidly from S2 to S3, it means that there are obvious hidden dangers in the system and immediate measures need to be taken.

[0175] 2. State transition probability

[0176] Output result: state transition probability matrix A = {a ij}.

[0177] Determination method:

[0178] High transition probability: If a 12 or a23 High means there is a high probability of transitioning from a normal state or potential hazard state to a more serious hazard state, indicating the need for close monitoring.

[0179] Low transition probability: If a 13 Low means that the probability of going directly from a normal state to a state of obvious hidden danger is low, and hidden dangers usually develop gradually.

[0180] Specific operations:

[0181] Check the transition probability a 12 : High value (such as a 12 >0.5) indicates that it is easier to transfer from a normal state to a potential hazard state.

[0182] Check the transition probability a 23 : High value (such as a 23 >0.5) indicates that the potential hidden danger state is easily transferred to the obvious hidden danger state.

[0183] Check the transition probability a 13 : Low value (such as a 13 <0.1) indicates that the probability of directly transferring from the normal state to the obvious hidden danger state is low.

[0184] 3. Observation Probability

[0185] Output result: Observation probability matrix B = {b j (o t ′)}.

[0186] Determination method:

[0187] High observation probability: In the explicit hidden danger state S3, the probability of observing abnormally high current is high, indicating that it is in a dangerous state.

[0188] Low Observation Probability: In normal state S1, the probability of observing abnormally high current is low, indicating normal operation.

[0189] Specific operations:

[0190] Check the observation probability b 3 (o t ′): high value (such as b 3 (o t ′)>0.5) indicates that the probability of observing high current in the explicit hidden danger state is high, and there may be danger.

[0191] Check the observation probability b 1 (o t ′): A low value (such as b1(ot′) < 0.1) indicates that the probability of observing high current under normal conditions is low and the operation is normal.

[0192] 4. Initial state probability

[0193] Output result: Initial state probability vector π = {π i}.

[0194] Determination method:

[0195] High initial probability: high π 1 It means that the situation is likely to be normal at the initial moment and hidden dangers may accumulate gradually.

[0196] Low initial probability: low π 3 It means that the probability of directly entering the state of obvious hidden danger at the initial moment is low, and hidden dangers usually develop gradually.

[0197] Specific operations:

[0198] Check the initial probability π 1 : High value (such as π 1 >0.8) indicates that the initial state is likely to be normal.

[0199] Check the initial probability π 3 : Low value (such as π 3 <0.1) indicates that the probability of directly entering the explicit hidden danger state at the initial moment is low.

[0200] 5. Predict future states

[0201] Output result: probability distribution of state at future time P(O T+1 ,q T+1 =S i |O).

[0202] Determination method:

[0203] High future state probability: If the explicit hidden danger state S 3 The probability is high, and there may be obvious hidden dangers in the future, which need to be prevented in advance.

[0204] Low future state probability: If the normal state S 1 The probability of it remaining normal in the future is low, and we need to pay attention to the development of hidden dangers.

[0205] Specific operations:

[0206] Check the future state probability P(O T+1 ,q T+1 =S 3 ∣O): A high value (such as P>0.5) indicates that there may be obvious hidden dangers in the future and prevention is needed.

[0207] Check the future state probability P(O T+1 ,q T+1 =S 1∣O): A low value (such as P<0.2) indicates that the probability of maintaining a normal state in the future is low and attention should be paid to hidden dangers.

[0208] In summary, the present invention achieves comprehensive coverage and accurate identification of hidden dangers in transmission lines by monitoring the multi-environmental conditions of traveling wave current and power frequency voltage data. Advanced signal processing algorithms such as wavelet transform and Fourier transform are used to extract time domain and frequency domain features, fully explore and utilize traveling wave current data, and improve the accuracy of hidden danger identification. The hidden Markov model is used to model time series data, and the early development of hidden dangers can be identified and warned through the analysis of state transition and observation probability, effectively preventing further deterioration of hidden dangers. Combined with the state transition and future trend prediction capabilities of the hidden Markov model, the development trend of hidden dangers in transmission lines can be predicted in a timely manner, helping to take preventive measures in advance to ensure the safety and stability of the transmission system.

[0209] The above is a schematic scheme of a method for identifying hidden dangers in a power transmission line of this embodiment. It should be noted that the technical scheme of the system for identifying hidden dangers in a power transmission line and the technical scheme of the method for identifying hidden dangers in a power transmission line belong to the same concept, and the details of the technical scheme of the system for identifying hidden dangers in a power transmission line that are not described in detail in this embodiment can be found in the description of the technical scheme of the method for identifying hidden dangers in a power transmission line.

[0210] This embodiment also provides a transmission line hidden danger identification system, including:

[0211] Acquisition module, extraction and analysis module, analysis module;

[0212] The acquisition module is used to collect the traveling wave current and power frequency voltage data during the simulation test of a specific voltage level, and record the discharge distance corresponding to the traveling wave current;

[0213] The extraction and analysis module is used to extract the time domain and frequency domain features of the traveling current waveform, construct a data matrix and analyze the hidden danger features;

[0214] The analysis module is used to perform time series analysis on the development of hidden dangers and identify early indicators of hidden danger development.

[0215] This embodiment further provides a computing device, which is applicable to the case of identifying hidden dangers of power transmission lines, and includes:

[0216] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method for identifying hidden dangers of transmission lines as proposed in the above embodiment.

[0217] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for identifying hidden dangers of a power transmission line as proposed in the above embodiment is implemented.

[0218] The storage medium proposed in this embodiment and the method for identifying hidden dangers of transmission lines proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0219] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0220] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for identifying hidden dangers of a power transmission line, characterized in that: include, Collect the traveling wave current and power frequency voltage data during the simulation test of a specific voltage level, and record the discharge distance corresponding to the traveling wave current; Extract the time domain and frequency domain features of the traveling current waveform, construct a data matrix and analyze the hidden danger characteristics; Conduct time series analysis on hidden danger development and identify early indicators of hidden danger development.

2. The method for identifying hidden dangers of a power transmission line according to claim 1, characterized in that: The recording of the discharge distance corresponding to the traveling wave current includes determining the critical discharge distances and the time points of different hidden danger types in combination with an ultraviolet imager.

3. The method for identifying hidden dangers of power transmission lines according to claim 2, characterized in that: The time domain and frequency domain feature extraction of the traveling wave current waveform includes: Pre-process the collected traveling current waveform to remove noise and interference; Apply wavelet transform technology to extract the subtle change characteristics of the traveling current waveform in the time domain; The Fourier transform method is used to obtain the distribution characteristics of the traveling wave current waveform in the frequency domain; The extracted time domain and frequency domain features are normalized; The constructing of the data matrix comprises: The extracted time domain features and frequency domain features are fused to form a comprehensive feature vector; Construct a data matrix containing multiple eigenvectors.

4. The method for identifying hidden dangers of power transmission lines according to claim 3, characterized in that: The analysis of hidden danger characteristics includes: The data matrix is ​​processed for dimensionality reduction to reduce the data dimension and extract the main components, which can be expressed as: Among them, Z is the feature matrix after dimensionality reduction, W is the feature matrix, X is the original data matrix, α and β are adjustment parameters, σ i is the standard deviation of the data sample, g(x) is a weighted kernel function used to enhance feature extraction; Furthermore, the kernel function g(x) is expressed as: Where x is the input data point; x j is the center point of the kernel function; w j is the kernel function weight; σ is the width parameter of the kernel function; M is the number of kernel function center points; Classify the feature vector after dimensionality reduction and identify the key features of the hidden danger type, which can be expressed as: Among them, S w is the intra-class scatter matrix, S b is the inter-class scatter matrix, x i is the data sample, C is the number of categories, N c is the number of samples of class c, μ c is the class mean, γ and δ are adjustment parameters, σ represents the overall standard deviation, f(x i ) nonlinear mapping function, h(μ c ,μ) is the nonlinear mapping function, J is the matrix transpose; Furthermore, f(x i ) and h(μ c ,μ) are respectively expressed as: f(x i )=tanh(ax i +b) Among them, a and b are the parameters of the mapping function; μ c is the class mean; μ is the overall mean; δ is the smoothing parameter to prevent the denominator from being zero; According to the analysis results, the main components and key features are extracted, and the characteristic models of different hidden danger types are established, which are expressed as: Among them, M is the feature model, λ i is the weight of the PCA principal component, z i is the feature after dimensionality reduction, η j is the weight of the LDA key feature, s j Key features identified by LDA.

5. The method for identifying hidden dangers of power transmission lines according to claim 4, characterized in that: The time series analysis of hidden danger development and identification of early indicators of hidden danger development include modeling the time series data of traveling wave current through a hidden Markov model to predict the trend and timing of hidden danger development.

6. The method for identifying hidden dangers of power transmission lines according to claim 5, characterized in that: The modeling includes: Arrange the traveling wave current data in chronological order to form a time series data set O = {o1, o2, ..., o T },o t represents the observed value at time t, and T represents the total time length; Standardize the data, calculate the mean and standard deviation of each observation, and obtain the standardized data ot′; Define the state space and observation sequence respectively, expressed as: S={S1,S2,…,S N } O={o1′,o2′,…,o T ′} Among them, S i represents the i-th state, there are N possible states in total; ot′ is the standardized observation value; Initialize the state probability vector, expressed as: Among them, π i Indicates that the system is in state S at time t = 1 i The probability of i is the basic initial state probability parameter, κ is the adjustment parameter, and λ is the frequency; Define the state transition probability matrix A = {a ij }, where a ij Indicates that from state S i Transfer to state S j The probability is expressed as: Among them, α ij is the basic transition probability parameter, β is the adjustment parameter, ω is the frequency, and φ is the phase; Define the observation probability matrix B = {b j (o t ′)}, where b j (o t ′) indicates that in state S j Observed below o t The probability of ′ is expressed as: Among them, μ j For state S j The observed mean, σ j For state S j The observed standard deviation, γ is the adjustment parameter, θ is the frequency, and δ is the phase; Perform model training and parameter estimation, which includes, Calculate the probability of the observed sequence using the forward algorithm; Use the Viterbi algorithm to find the most likely sequence of states; According to the results of the forward and Viterbi algorithms, the expectation-maximization algorithm is used to iteratively update the parameters of π, A, and B; Predict the trend and timing of hidden danger development, including: Given the current observation sequence O = {o1′, o2′, …, o T ′}, the observation probability at future moments under different states is calculated by the forward algorithm, which is expressed as: Predict the trend and time series of hidden danger development, and analyze the periodicity and nonlinear characteristics of hidden danger changes based on the time dependence term γcos(θt+δ) in the observation probability matrix B.

7. The method for identifying hidden dangers of a power transmission line according to claim 6, characterized in that: The method of calculating the probability of the observation sequence using the forward algorithm includes: Initialize the forward variable α1(i): α1(i)=π i b i (o1′) Recursively calculate the forward variable αt(i): The method of using the Viterbi algorithm to find the most likely state sequence includes: Initialize the Viterbi variable δ1(i): δ1(i)=π i b i (o1′) Recursively calculate the Viterbi variable δt(i): d t (i)=max 1≤j≤N [d t-1 (j)A ji ]B i (the t ′).

8. A system using the method for identifying hidden dangers of power transmission lines as claimed in any one of claims 1 to 7, characterized in that: It includes acquisition module, extraction and analysis module and analysis module; The acquisition module is used to collect the traveling wave current and power frequency voltage data during the simulation test of a specific voltage level, and record the discharge distance corresponding to the traveling wave current; The extraction and analysis module is used to extract the time domain and frequency domain features of the traveling current waveform, construct a data matrix and analyze the hidden danger features; The analysis module is used to perform time series analysis on the development of hidden dangers and identify early indicators of hidden danger development.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the power transmission line hidden danger identification method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the power transmission line hidden danger identification method described in any one of claims 1 to 7 are implemented.

Images