CN119249136A - Early warning method for transmission line icing based on optical fiber vibration data hierarchical context - Google Patents
Early warning method for transmission line icing based on optical fiber vibration data hierarchical context Download PDFInfo
- Publication number
- CN119249136A CN119249136A CN202411787705.2A CN202411787705A CN119249136A CN 119249136 A CN119249136 A CN 119249136A CN 202411787705 A CN202411787705 A CN 202411787705A CN 119249136 A CN119249136 A CN 119249136A
- Authority
- CN
- China
- Prior art keywords
- optical fiber
- fiber vibration
- vibration data
- list
- time domain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H9/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
- G01H9/004—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means using fibre optic sensors
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Optical Communication System (AREA)
Abstract
本发明提供一种基于光纤震动数据层级上下文的输电线初期覆冰预警方法，包括：在输电线路上安装光纤和相位敏感光时域反射计；建立时域指标表达模块；建立光纤震动数据层级上下文特性抽取模块；建立分层次覆冰决策神经网模型；利用光纤震动数据层级上下文特性抽取模块和训练完成的分层次覆冰决策神经网模型，对采集到的4秒钟时长的光纤震动数据进行分析，判断是否出现输电线初期覆冰情况，如果是，则进行预警。通过本发明的分层次结构，可以描述光纤震动数据不同周期之间的变化过程，可以更加有效的进行输电线初期覆冰预警，提高电网巡检管理的效率。
The present invention provides a method for early warning of transmission line icing based on optical fiber vibration data hierarchical context, including: installing optical fiber and phase-sensitive optical time domain reflectometer on the transmission line; establishing a time domain index expression module; establishing an optical fiber vibration data hierarchical context feature extraction module; establishing a hierarchical icing decision neural network model; using the optical fiber vibration data hierarchical context feature extraction module and the trained hierarchical icing decision neural network model, the collected 4-second optical fiber vibration data is analyzed to determine whether the transmission line has early icing, and if so, an early warning is issued. Through the hierarchical structure of the present invention, the change process between different periods of optical fiber vibration data can be described, and early warning of transmission line icing can be more effectively performed, thereby improving the efficiency of power grid inspection management.
Description
Technical Field
The invention relates to the technical field of remote monitoring of transmission lines, in particular to an early-stage icing early-warning method of a transmission line based on the context of an optical fiber vibration data level.
Background
In northern cold areas, the icing of the power transmission line often occurs, and the power transmission line equipment is easy to damage, so that huge economic loss and safety accidents are caused, and the power transmission line needs to be treated as early as possible in the early stage of the thinner icing of the power transmission line. The optical fiber deployed on the power transmission line can rapidly collect vibration data of a corresponding line, and the vibration mode of the power transmission line can change after icing, so that the icing condition of the power transmission line is very necessary to be monitored remotely by utilizing the vibration data of the optical fiber, early warning and processing can be performed as soon as possible when the power transmission line is iced at the initial stage, and the management efficiency of a power grid maintenance enterprise can be remarkably improved.
The main method for collecting the initial icing of the transmission line by utilizing the optical fiber vibration data at present is to collect the optical fiber vibration data on the transmission line by utilizing a phase sensitive optical time domain reflectometer and analyze the optical fiber vibration data by utilizing an artificial intelligent model. The main method adopted at present comprises 1) directly processing by using an artificial intelligent model, wherein the frequency of optical fiber vibration data collected by a phase sensitive optical time domain reflectometer is higher, the generated relevant time-frequency attribute is more, and the processing mode is usually limited in precision because of fewer samples reflecting the initial icing state of a power transmission line, 2) relatively effective response to the problem of overhigh attribute dimension of the optical fiber vibration data by using a Boost or Bagging integrated classifier, but the characteristic change of the optical fiber vibration data is caused by different power transmission line lengths and assembly modes in a real application scene, so that the optical fiber vibration data obtained by a laboratory is slightly different from the actual optical fiber vibration data, and the experimental result of the mode in the laboratory is usually better, but in practical application, the method may be fitted with a specific mode to cause prediction failure. Especially in the case of initial icing of a power transmission line, the mode is similar to the non-icing state due to the low icing thickness, and the effects of the two methods are poor, and 3) a complex deep neural network model is utilized to distinguish the core difference between the initial icing and the non-icing, and the biggest problem of the mode is that the number of the samples of the power transmission line icing is small and insufficient to support the full training of a large neural network, so that the excessive fitting phenomenon is caused, and the early warning precision cannot meet the requirement.
Therefore, a method is required to be provided, which can adapt to the difference between the optical fiber vibration data of the thinner icing and the non-icing in the initial stage of the power transmission line, adapt to the difference caused by the construction of different power transmission lines, and accurately and early warn the initial icing of the power transmission line.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a transmission line initial icing early warning method based on the context of an optical fiber vibration data level, which constructs an optical fiber vibration data level context characteristic extraction module and a hierarchical icing decision neural network model, and (3) establishing the characteristics of the optical fiber vibration data in a layered manner and finding the difference between the optical fiber vibration data of which the ice is thinly covered and not covered at the initial stage of the power transmission line, so as to realize early warning of the ice covering at the initial stage of the power transmission line.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
The invention provides an early-stage icing early-warning method for a power transmission line based on the context of an optical fiber vibration data level, which comprises the following steps:
The method comprises the steps of S1, installing an optical fiber and a phase sensitive optical time domain reflectometer on a power transmission line, wherein the phase sensitive optical time domain reflectometer is used for collecting optical fiber vibration data, collecting historical data formed by the optical fiber vibration data and the icing condition of the power transmission line, and storing the historical data in the icing condition of the power transmission line into an icing condition historical data table VbList, obtaining the number of elements of the icing condition historical data table VbNum, initializing an optical fiber vibration data level context characteristic list CJList;
s2, establishing a time domain index expression module FEModel, wherein the input of the time domain index expression module FEModel is a time domain index expression input array FEInput, and the output of the time domain index expression module FEModel is a time domain index expression output array FEOutput;
S3, establishing an optical fiber vibration data level context characteristic extraction module TXModel, wherein the input of the optical fiber vibration data level context characteristic extraction module TXModel is a floating point array TXInput with 4000 elements, the optical fiber vibration data level context characteristic extraction module TXModel processes the floating point array TXInput by utilizing the time domain index expression module FEModel, and the output of the optical fiber vibration data level context characteristic extraction module TXModel is an optical fiber vibration data level context characteristic output result TXOutput;
S4, utilizing the optical fiber vibration data level context characteristic extraction module TXModel to process all contents of the ice-covering condition form history data table VbList to obtain optical fiber vibration data level context characteristics, and storing the obtained optical fiber vibration data level context characteristics in the optical fiber vibration data level context characteristics list CJList;
S5, establishing a hierarchical icing decision neural network model CCNN, training the hierarchical icing decision neural network model CCNN by utilizing the optical fiber vibration data level context characteristic list CJList to obtain a trained hierarchical icing decision neural network model CCNN;
And S6, during actual monitoring, acquiring optical fiber vibration data Test with the duration of 4 seconds by using the phase sensitive optical time domain reflectometer, analyzing the acquired optical fiber vibration data Test with the duration of 4 seconds by using the optical fiber vibration data hierarchy context characteristic extraction module TXModel and the trained hierarchical icing decision neural network model CCNN, and judging whether the initial icing condition of the power transmission line occurs or not, and if so, carrying out early warning.
Preferably, S1 is specifically:
S101, installing an optical fiber and a phase sensitive optical time domain reflectometer on a power transmission line;
the phase sensitive optical time domain reflectometer collects optical fiber vibration data at 1000 hertz, collects historical data formed by the optical fiber vibration data and the icing condition form of the transmission line, and stores the historical data in the icing condition form historical data table VbList, wherein each element of the icing condition form historical data table VbList has 2 fields:
Optical fiber vibration data array VbData the optical fiber vibration data array VbData is a floating point array with 4000 elements, and because the acquisition frequency of the phase sensitive optical time domain reflectometer is 1000 Hz, the optical fiber vibration data array VbData corresponds to optical fiber vibration data with the duration of 4 seconds;
Whether thinner icing VbObserved occurs or not is an integer variable, 1 indicates that thinner icing occurs, at this time, the thickness of the icing of the power transmission line is more than 0 and less than 5mm, and 0 indicates that no icing occurs;
s102, the number VbNum of elements of the ice-over condition form history data table=the number of elements of the ice-over condition form history data table VbList;
S103, establishing a fiber vibration data level context characteristic list CJList which is initially an empty list, wherein each element structure of the fiber vibration data level context characteristic list CJList comprises 6 fields:
the first-level context characteristic Vb1 of the optical fiber vibration characteristic is a list of 40 elements, and each element in the list is a vector of 10 dimensions;
the second-level context characteristic of the optical fiber vibration characteristic is Vb2, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
The third-level context characteristic of the vibration characteristic of the optical fiber is Vb3, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
the fourth-level context characteristic of the vibration characteristic of the optical fiber is Vb4, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
The fifth level context characteristic of the vibration characteristic of the optical fiber is Vb5, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
The icing decision FBDecison of the fiber vibration data level context characteristic list CJList is an integer variable, expressing whether icing occurs;
S104, constructing a fiber vibration data level context characteristic list counter CJCounter =1;
S105, one element CJListItem of the fiber vibration data hierarchy context characteristics list CJList is established, and the element CJListItem has 6 fields:
the first-level context characteristic Vb1 of the optical fiber vibration characteristic is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
the second-level context characteristic of the optical fiber vibration characteristic is that the optical fiber vibration characteristic is a list of 40 elements, each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
The third-level context characteristic of the optical fiber vibration characteristic is Vb3, which is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
the fourth-level context characteristic of the vibration characteristic of the optical fiber is Vb4, which is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
The fifth level context characteristic of the optical fiber vibration characteristic is Vb5, which is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
Icing decision FBDecison = 0;
S106, ending the step S1.
Preferably, S2 is specifically:
S201, a time domain index expression module FEModel is established, wherein the input of the time domain index expression module FEModel is a time domain index expression input array FEInput, and the time domain index expression input array FEInput is an array of 100 elements;
s202, performing the following formula calculation on the time domain index expression input array FEInput:
(1) Time domain standard deviation index:
standard deviation sigma, the degree of dispersion of the measured data, and the calculation formula is as follows:
wherein, the average mu calculation formula is as follows:
Wherein: N=100, representing the number of elements in the time domain indicator representation input array FEInput;
(2) Time domain peak-to-peak index:
the peak-to-peak value PP represents the extremum range in the time domain indicator expression input array FEInput, and the calculation formula is:
Wherein: representing the maximum value of an element in the time domain indicator representation input array FEInput; representing the element minimum in the time domain indicator representation input array FEInput;
(3) Time domain skewness index:
Degree of deviation Measuring the symmetry of data distribution, wherein the calculation formula is as follows:
(4) Time domain margin index:
the margin R represents the relative size of peak-to-peak value and average value, and the calculation formula is as follows:
(5) Time domain kurtosis index:
Kurtosis of Describing the sharpness of data distribution, the calculation formula is:
(6) Time domain maximum index:
the maximum value Max represents the maximum element in the time domain index expression input array FEInput, and the calculation formula is:
(7) Time domain minimum index:
the minimum value Min represents the minimum element in the time domain index expression input array FEInput, and the calculation formula is:
(8) Time domain pulse index:
The pulse index P represents the pulse intensity of the signal, and is defined as:
Wherein: Representing the absolute value of the ith element in the time domain indicator representation input array FEInput;
(9) Time domain waveform value index:
Waveform value The integral fluctuation of the signal is measured, and the calculation formula is as follows:
(10) Time domain energy index:
the energy E represents the sum of squares of all elements in the time domain index expression input array FEInput, and the calculation formula is:
s203, establishing a time domain index expression output array FEOutput as an array of 10 elements;
S204, sequentially storing 10 index results calculated in the S202 into a time domain index expression output array FEOutput;
s205, the time domain indicator expression output array FEOutput is output as the result of the time domain indicator expression module FEModel.
Preferably, S3 is specifically:
S301, establishing an optical fiber vibration data level context characteristic extraction module TXModel, wherein the input of the optical fiber vibration data level context characteristic extraction module TXModel is a floating point array TXInput with 4000 elements;
S302, the first counter TXCounter1 =1 of the context characteristic extraction module of the optical fiber vibration data level is made to construct a list variable TXList =empty list of the context characteristic extraction module of the optical fiber vibration data level;
S303, a first temporary variable TXTemp of the context characteristic extraction module of the optical fiber vibration data level = the (TXCounter 1-1) x 100 th to (TXCounter 1-1) x 100+99 th elements of the extracted floating point array TXInput, totaling 100 elements;
s304, the fiber vibration data level context characteristics extraction module second temporary variable TXTemp2:
Processing the first temporary storage variable TXTemp1 of the optical fiber vibration data level context characteristic extraction module by using the time domain index expression module FEModel, wherein the time domain index expression input array FEInput =the first temporary storage variable TXTemp1 of the optical fiber vibration data level context characteristic extraction module of the time domain index expression module FEModel to obtain a time domain index expression output array FEOutput;
S305, adding the second temporary variable TXTemp of the context characteristic extraction module of the optical fiber vibration data level to the list variable TXList of the context characteristic extraction module of the optical fiber vibration data level;
S306,TXCounter1=TXCounter1+1;
s307, if TXCounter is less than or equal to 40, turning to S303, otherwise turning to S308;
s308, establish a fiber vibration data hierarchy context characteristic output result TXOutput, where the structure of the fiber vibration data hierarchy context characteristic output result TXOutput is the same as one element structure of the fiber vibration data hierarchy context characteristic list CJList, and includes 6 fields:
The Vb1 field of the fiber vibration data level context characteristic output result TXOutput = fiber vibration data level context characteristic extraction module list variable TXList, specifically, the Vb1 field of the fiber vibration data level context characteristic output result TXOutput is a list of 40 elements, each element in the list is a 10-dimensional vector;
The Vb2 field of the fiber vibration data level contextual characteristics output result TXOutput = create a new list with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb1 field and delete an element at the end of its Vb1 field;
the Vb3 field of the fiber vibration data level contextual characteristics output result TXOutput = create a new list with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb2 field and delete an element at the end of its Vb2 field;
The Vb4 field of the fiber vibration data level contextual characteristics output result TXOutput = create a new list with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb3 field and delete an element at the end of its Vb3 field;
the Vb5 field of the fiber vibration data level contextual characteristics output result TXOutput = a new list is created with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb4 field and deleting an element at the end of its Vb4 field;
An icing decision FBDecison field of the context characteristic output result TXOutput of the optical fiber vibration data level is an integer variable and expresses whether icing occurs or not;
s309, updating each field of the fiber vibration data hierarchy context characteristic output result TXOutput established in S308:
Vb2 field of fiber vibration data level context characteristic output result TXOutput = Vb2 field of fiber vibration data level context characteristic extraction module list variable TXList-fiber vibration data level context characteristic output result TXOutput;
vb3 field of fiber vibration data level context characteristic output result TXOutput = fiber vibration data level context characteristic extraction module list variable TXList-Vb 3 field of fiber vibration data level context characteristic output result TXOutput;
vb4 field of fiber vibration data level context characteristic output result TXOutput = Vb4 field of fiber vibration data level context characteristic extraction module list variable TXList-fiber vibration data level context characteristic output result TXOutput;
Vb5 field of fiber vibration data level context characteristic output result TXOutput = fiber vibration data level context characteristic extraction module list variable TXList-Vb 5 field of fiber vibration data level context characteristic output result TXOutput;
Icing decision FBDecison field = 0 for fiber vibration data level contextual characteristic output result TXOutput;
S310, the fiber vibration data level context characteristic output result TXOutput is output as the result of the fiber vibration data level context characteristic extraction module TXModel.
Preferably, S4 is specifically:
s401, result processing counter CLCounter =1;
s402, the result processes the first temporary variable CLTemp1:
The CLCounter th element of the history data table VbList in the ice-over condition form is fetched and assigned to the result processing first temporary variable CLTemp;
s403, the result processing second temporary variable CLTemp2:
The optical fiber vibration data level context characteristic extraction module TXModel is utilized to process, the floating point array TXInput input to the optical fiber vibration data level context characteristic extraction module TXModel processes the first temporary variable CLTemp1 for the result, obtains the optical fiber vibration data level context characteristic output result TXOutput through processing, and assigns the result to the result processing second temporary variable CLTemp;
s404, assigning the result processing second temporary variable CLTemp to the CLCounter element of the context characteristic list CJList of the optical fiber vibration data hierarchy;
S405, taking out whether a thinner icing VbObserved field appears on the CLCounter th element of the icing condition form history data table VbList, and assigning the thinner icing decision FBDecison field to the CLCounter th element of the optical fiber vibration data level context characteristic list CJList;
S406,CLCounter=CLCounter+1;
s407, if CLCounter is less than or equal to VbNum, turning to S402, otherwise turning to S408;
s408, the processing procedure of the S4 step is ended.
Preferably, S5 is specifically:
S501, a hierarchical icing decision neural network model CCNN is established, which comprises the following structures:
the method comprises the steps of outputting an output array of 100 elements through an LSTM layer by using 5 input branches, combining the outputs into a 100X 5 feature map after the 5 input branches are processed, and forming a decision by using 2 convolution layers, 1 max pooling layer and a three-time MLP network for the feature map;
S502, five input branches of the hierarchical icing decision neural network model CCNN are respectively connected into fields vb1, vb2, vb3, vb4 and vb5 of the context characteristic list CJList of the optical fiber vibration data level, and decision output of the hierarchical icing decision neural network model CCNN is connected into a field of icing decision FBDecison of the context characteristic list CJList of the optical fiber vibration data level;
S503, training a hierarchical icing decision neural network model CCNN by utilizing the data of the context characteristic list CJList of the optical fiber vibration data level, so that the hierarchical icing decision neural network model CCNN has decision capability;
s504, the processing procedure of the step S5 is ended.
Preferably, S6 is specifically:
s601, acquiring optical fiber vibration data Test with the duration of 4 seconds by using a phase sensitive optical time domain reflectometer;
s602, a first temporary variable PDTemp for determining:
Processing by using the optical fiber vibration data level context characteristic extraction module TXModel, inputting the floating point array TXInput of the optical fiber vibration data level context characteristic extraction module TXModel as the optical fiber vibration data Test, obtaining the optical fiber vibration data level context characteristic output result TXOutput of the optical fiber vibration data level context characteristic extraction module TXModel, and assigning the output result to the first temporary variable PDTemp1 for judgment;
S603, inputting the first temporary variable PDTemp for judgment to the hierarchical icing decision neural network model CCNN, and respectively accessing the vb1, vb2, vb3, vb4 and vb5 fields of the first temporary variable PDTemp for judgment into five input branches of the hierarchical icing decision neural network model CCNN;
S604, a second temporary variable PDTemp for determining:
the decision output of the hierarchical icing decision neural network model CCNN is assigned to a second temporary variable PDTemp for judgment;
s605, if the second temporary variable PDTemp for judgment is equal to 0, turning to S606, otherwise turning to S607;
s606, outputting that the initial icing condition of the power transmission line does not occur, and turning to S608;
s607, representing that the initial icing condition of the power transmission line appears, carrying out early warning, and turning to S608;
S608, end.
The early-stage icing early-warning method for the power transmission line based on the optical fiber vibration data level context has the following advantages:
The invention can describe the change process between different periods of the optical fiber vibration data through the hierarchical structure, on one hand, the neural network can find out the corresponding icing early warning of a specific level, adapt to the characteristic change of the optical fiber vibration data caused by different construction modes and assembly modes, on the other hand, the difference calculated by a formula between the levels can be independent of samples, the difference between different periods of the optical fiber vibration data can be expressed more clearly, and the invention brings help for the situation of fewer initial icing samples of the transmission line. The method and the system can more effectively perform early-stage icing early warning of the power transmission line and improve the efficiency of power grid inspection management.
Drawings
Fig. 1 is a schematic flow chart of a method for early warning ice coating on a transmission line based on a fiber vibration data level context.
Detailed Description
In order to make the technical problems, technical schemes and beneficial effects solved by the invention more clear, the invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The invention provides a transmission line initial icing early warning method based on the context of an optical fiber vibration data level, which constructs an optical fiber vibration data level context characteristic extraction module and a hierarchical icing decision neural network model, and (3) establishing the characteristics of the optical fiber vibration data in a layered manner and finding the difference between the optical fiber vibration data of which the ice is thinly covered and not covered at the initial stage of the power transmission line, so as to realize early warning of the ice covering at the initial stage of the power transmission line.
Referring to fig. 1, the invention provides a transmission line initial icing early warning method based on the context of an optical fiber vibration data level, which comprises the following steps:
The method comprises the steps of S1, installing an optical fiber and a phase sensitive optical time domain reflectometer on a power transmission line, wherein the phase sensitive optical time domain reflectometer is used for collecting optical fiber vibration data, collecting historical data formed by the optical fiber vibration data and the icing condition of the power transmission line, and storing the historical data in the icing condition of the power transmission line into an icing condition historical data table VbList, obtaining the number of elements of the icing condition historical data table VbNum, initializing an optical fiber vibration data level context characteristic list CJList;
s1 specifically comprises the following steps:
S101, installing an optical fiber and a phase sensitive optical time domain reflectometer on a power transmission line;
the phase sensitive optical time domain reflectometer collects optical fiber vibration data at 1000 hertz, collects historical data formed by the optical fiber vibration data and the icing condition form of the transmission line, and stores the historical data in the icing condition form historical data table VbList, wherein each element of the icing condition form historical data table VbList has 2 fields:
Optical fiber vibration data array VbData the optical fiber vibration data array VbData is a floating point array with 4000 elements, and because the acquisition frequency of the phase sensitive optical time domain reflectometer is 1000 Hz, the optical fiber vibration data array VbData corresponds to optical fiber vibration data with the duration of 4 seconds;
Whether thinner icing VbObserved occurs or not is an integer variable, 1 indicates that thinner icing occurs, at this time, the thickness of the icing of the power transmission line is more than 0 and less than 5mm, and 0 indicates that no icing occurs;
s102, the number VbNum of elements of the ice-over condition form history data table=the number of elements of the ice-over condition form history data table VbList;
S103, establishing a fiber vibration data level context characteristic list CJList which is initially an empty list, wherein each element structure of the fiber vibration data level context characteristic list CJList comprises 6 fields:
the first-level context characteristic Vb1 of the optical fiber vibration characteristic is a list of 40 elements, and each element in the list is a vector of 10 dimensions;
the second-level context characteristic of the optical fiber vibration characteristic is Vb2, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
The third-level context characteristic of the vibration characteristic of the optical fiber is Vb3, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
the fourth-level context characteristic of the vibration characteristic of the optical fiber is Vb4, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
The fifth level context characteristic of the vibration characteristic of the optical fiber is Vb5, which is a list of 40 elements, wherein each element in the list is a vector of 10 dimensions;
The icing decision FBDecison of the fiber vibration data level context characteristic list CJList is an integer variable, expressing whether icing occurs;
S104, constructing a fiber vibration data level context characteristic list counter CJCounter =1;
S105, one element CJListItem of the fiber vibration data hierarchy context characteristics list CJList is established, and the element CJListItem has 6 fields:
the first-level context characteristic Vb1 of the optical fiber vibration characteristic is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
the second-level context characteristic of the optical fiber vibration characteristic is that the optical fiber vibration characteristic is a list of 40 elements, each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
The third-level context characteristic of the optical fiber vibration characteristic is Vb3, which is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
the fourth-level context characteristic of the vibration characteristic of the optical fiber is Vb4, which is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
The fifth level context characteristic of the optical fiber vibration characteristic is Vb5, which is a list of 40 elements, wherein each element in the list is a 10-dimensional vector, and the interior of the vector is 0;
Icing decision FBDecison = 0;
S106, ending the step S1.
S2, establishing a time domain index expression module FEModel, wherein the input of the time domain index expression module FEModel is a time domain index expression input array FEInput, and the output of the time domain index expression module FEModel is a time domain index expression output array FEOutput;
S2 specifically comprises the following steps:
S201, a time domain index expression module FEModel is established, wherein the input of the time domain index expression module FEModel is a time domain index expression input array FEInput, and the time domain index expression input array FEInput is an array of 100 elements;
s202, performing the following formula calculation on the time domain index expression input array FEInput:
(1) Time domain standard deviation index:
standard deviation sigma, the degree of dispersion of the measured data, and the calculation formula is as follows:
wherein, the average mu calculation formula is as follows:
Wherein: N=100, representing the number of elements in the time domain indicator representation input array FEInput;
(2) Time domain peak-to-peak index:
the peak-to-peak value PP represents the extremum range in the time domain indicator expression input array FEInput, and the calculation formula is:
Wherein: representing the maximum value of an element in the time domain indicator representation input array FEInput; representing the element minimum in the time domain indicator representation input array FEInput;
(3) Time domain skewness index:
Degree of deviation Measuring the symmetry of data distribution, wherein the calculation formula is as follows:
(4) Time domain margin index:
the margin R represents the relative size of peak-to-peak value and average value, and the calculation formula is as follows:
(5) Time domain kurtosis index:
Kurtosis of Describing the sharpness of data distribution, the calculation formula is:
(6) Time domain maximum index:
the maximum value Max represents the maximum element in the time domain index expression input array FEInput, and the calculation formula is:
(7) Time domain minimum index:
the minimum value Min represents the minimum element in the time domain index expression input array FEInput, and the calculation formula is:
(8) Time domain pulse index:
The pulse index P represents the pulse intensity of the signal, and is defined as:
Wherein: Representing the absolute value of the ith element in the time domain indicator representation input array FEInput;
(9) Time domain waveform value index:
Waveform value The integral fluctuation of the signal is measured, and the calculation formula is as follows:
(10) Time domain energy index:
the energy E represents the sum of squares of all elements in the time domain index expression input array FEInput, and the calculation formula is:
s203, establishing a time domain index expression output array FEOutput as an array of 10 elements;
S204, sequentially storing 10 index results calculated in the S202 into a time domain index expression output array FEOutput;
s205, the time domain indicator expression output array FEOutput is output as the result of the time domain indicator expression module FEModel.
S3, establishing an optical fiber vibration data level context characteristic extraction module TXModel, wherein the input of the optical fiber vibration data level context characteristic extraction module TXModel is a floating point array TXInput with 4000 elements, the optical fiber vibration data level context characteristic extraction module TXModel processes the floating point array TXInput by utilizing the time domain index expression module FEModel, and the output of the optical fiber vibration data level context characteristic extraction module TXModel is an optical fiber vibration data level context characteristic output result TXOutput;
s3 specifically comprises the following steps:
S301, establishing an optical fiber vibration data level context characteristic extraction module TXModel, wherein the input of the optical fiber vibration data level context characteristic extraction module TXModel is a floating point array TXInput with 4000 elements;
S302, the first counter TXCounter1 =1 of the context characteristic extraction module of the optical fiber vibration data level is made to construct a list variable TXList =empty list of the context characteristic extraction module of the optical fiber vibration data level;
S303, a first temporary variable TXTemp of the context characteristic extraction module of the optical fiber vibration data level = the (TXCounter 1-1) x 100 th to (TXCounter 1-1) x 100+99 th elements of the extracted floating point array TXInput, totaling 100 elements;
s304, the fiber vibration data level context characteristics extraction module second temporary variable TXTemp2:
Processing the first temporary storage variable TXTemp1 of the optical fiber vibration data level context characteristic extraction module by using the time domain index expression module FEModel, wherein the time domain index expression input array FEInput =the first temporary storage variable TXTemp1 of the optical fiber vibration data level context characteristic extraction module of the time domain index expression module FEModel to obtain a time domain index expression output array FEOutput;
S305, adding the second temporary variable TXTemp of the context characteristic extraction module of the optical fiber vibration data level to the list variable TXList of the context characteristic extraction module of the optical fiber vibration data level;
S306,TXCounter1=TXCounter1+1;
s307, if TXCounter is less than or equal to 40, turning to S303, otherwise turning to S308;
s308, establish a fiber vibration data hierarchy context characteristic output result TXOutput, where the structure of the fiber vibration data hierarchy context characteristic output result TXOutput is the same as one element structure of the fiber vibration data hierarchy context characteristic list CJList, and includes 6 fields:
The Vb1 field of the fiber vibration data level context characteristic output result TXOutput = fiber vibration data level context characteristic extraction module list variable TXList, specifically, the Vb1 field of the fiber vibration data level context characteristic output result TXOutput is a list of 40 elements, each element in the list is a 10-dimensional vector;
The Vb2 field of the fiber vibration data level contextual characteristics output result TXOutput = create a new list with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb1 field and delete an element at the end of its Vb1 field;
the Vb3 field of the fiber vibration data level contextual characteristics output result TXOutput = create a new list with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb2 field and delete an element at the end of its Vb2 field;
The Vb4 field of the fiber vibration data level contextual characteristics output result TXOutput = create a new list with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb3 field and delete an element at the end of its Vb3 field;
the Vb5 field of the fiber vibration data level contextual characteristics output result TXOutput = a new list is created with a value of inserting a 10-dimensional vector of all 0's before the first element of its Vb4 field and deleting an element at the end of its Vb4 field;
An icing decision FBDecison field of the context characteristic output result TXOutput of the optical fiber vibration data level is an integer variable and expresses whether icing occurs or not;
s309, updating each field of the fiber vibration data hierarchy context characteristic output result TXOutput established in S308:
Vb2 field of fiber vibration data level context characteristic output result TXOutput = Vb2 field of fiber vibration data level context characteristic extraction module list variable TXList-fiber vibration data level context characteristic output result TXOutput;
vb3 field of fiber vibration data level context characteristic output result TXOutput = fiber vibration data level context characteristic extraction module list variable TXList-Vb 3 field of fiber vibration data level context characteristic output result TXOutput;
vb4 field of fiber vibration data level context characteristic output result TXOutput = Vb4 field of fiber vibration data level context characteristic extraction module list variable TXList-fiber vibration data level context characteristic output result TXOutput;
Vb5 field of fiber vibration data level context characteristic output result TXOutput = fiber vibration data level context characteristic extraction module list variable TXList-Vb 5 field of fiber vibration data level context characteristic output result TXOutput;
Icing decision FBDecison field = 0 for fiber vibration data level contextual characteristic output result TXOutput;
S310, the fiber vibration data level context characteristic output result TXOutput is output as the result of the fiber vibration data level context characteristic extraction module TXModel.
S4, utilizing the optical fiber vibration data level context characteristic extraction module TXModel to process all contents of the ice-covering condition form history data table VbList to obtain optical fiber vibration data level context characteristics, and storing the obtained optical fiber vibration data level context characteristics in the optical fiber vibration data level context characteristics list CJList;
S4 specifically comprises the following steps:
s401, result processing counter CLCounter =1;
s402, the result processes the first temporary variable CLTemp1:
The CLCounter th element of the history data table VbList in the ice-over condition form is fetched and assigned to the result processing first temporary variable CLTemp;
s403, the result processing second temporary variable CLTemp2:
The optical fiber vibration data level context characteristic extraction module TXModel is utilized to process, the floating point array TXInput input to the optical fiber vibration data level context characteristic extraction module TXModel processes the first temporary variable CLTemp1 for the result, obtains the optical fiber vibration data level context characteristic output result TXOutput through processing, and assigns the result to the result processing second temporary variable CLTemp;
s404, assigning the result processing second temporary variable CLTemp to the CLCounter element of the context characteristic list CJList of the optical fiber vibration data hierarchy;
S405, taking out whether a thinner icing VbObserved field appears on the CLCounter th element of the icing condition form history data table VbList, and assigning the thinner icing decision FBDecison field to the CLCounter th element of the optical fiber vibration data level context characteristic list CJList;
S406,CLCounter=CLCounter+1;
s407, if CLCounter is less than or equal to VbNum, turning to S402, otherwise turning to S408;
s408, the processing procedure of the S4 step is ended.
S5, establishing a hierarchical icing decision neural network model CCNN, training the hierarchical icing decision neural network model CCNN by utilizing the optical fiber vibration data level context characteristic list CJList to obtain a trained hierarchical icing decision neural network model CCNN;
s5 specifically comprises the following steps:
S501, a hierarchical icing decision neural network model CCNN is established, which comprises the following structures:
the method comprises the steps of outputting an output array of 100 elements through an LSTM layer by using 5 input branches, combining the outputs into a 100X 5 feature map after the 5 input branches are processed, and forming a decision by using 2 convolution layers, 1 max pooling layer and a three-time MLP network for the feature map;
S502, five input branches of the hierarchical icing decision neural network model CCNN are respectively connected into fields vb1, vb2, vb3, vb4 and vb5 of the context characteristic list CJList of the optical fiber vibration data level, and decision output of the hierarchical icing decision neural network model CCNN is connected into a field of icing decision FBDecison of the context characteristic list CJList of the optical fiber vibration data level;
S503, training a hierarchical icing decision neural network model CCNN by utilizing the data of the context characteristic list CJList of the optical fiber vibration data level, so that the hierarchical icing decision neural network model CCNN has decision capability;
s504, the processing procedure of the step S5 is ended.
And S6, during actual monitoring, acquiring optical fiber vibration data Test with the duration of 4 seconds by using the phase sensitive optical time domain reflectometer, analyzing the acquired optical fiber vibration data Test with the duration of 4 seconds by using the optical fiber vibration data hierarchy context characteristic extraction module TXModel and the trained hierarchical icing decision neural network model CCNN, and judging whether the initial icing condition of the power transmission line occurs or not, and if so, carrying out early warning.
S6 is specifically as follows:
s601, acquiring optical fiber vibration data Test with the duration of 4 seconds by using a phase sensitive optical time domain reflectometer;
s602, a first temporary variable PDTemp for determining:
Processing by using the optical fiber vibration data level context characteristic extraction module TXModel, inputting the floating point array TXInput of the optical fiber vibration data level context characteristic extraction module TXModel as the optical fiber vibration data Test, obtaining the optical fiber vibration data level context characteristic output result TXOutput of the optical fiber vibration data level context characteristic extraction module TXModel, and assigning the output result to the first temporary variable PDTemp1 for judgment;
S603, inputting the first temporary variable PDTemp for judgment to the hierarchical icing decision neural network model CCNN, and respectively accessing the vb1, vb2, vb3, vb4 and vb5 fields of the first temporary variable PDTemp for judgment into five input branches of the hierarchical icing decision neural network model CCNN;
S604, a second temporary variable PDTemp for determining:
the decision output of the hierarchical icing decision neural network model CCNN is assigned to a second temporary variable PDTemp for judgment;
s605, if the second temporary variable PDTemp for judgment is equal to 0, turning to S606, otherwise turning to S607;
s606, outputting that the initial icing condition of the power transmission line does not occur, and turning to S608;
s607, representing that the initial icing condition of the power transmission line appears, carrying out early warning, and turning to S608;
S608, end.
In order to verify the effectiveness of the method, data of a transmission line in winter in Jilin areas are introduced, 1000 time period optical fiber vibration data are extracted, 100, 50 and 50 thinner initial icing samples are used as training samples, 50 are used as test samples, 900 non-icing samples, 450 are used as training samples and 450 are used as test samples), and the method is compared with the traditional method as follows:
| method name | Correct early warning number of 50 icing samples | 450 Number of error early warning of non-icing samples |
| The method of the invention | 50 | 1 |
| Traditional method 1 decision Tree | 25 | 12 |
| Traditional method 2 boosting+neural network | 28 | 8 |
| Traditional method 3 deep neural network | 37 | 6 |
From the table, the method is superior to other traditional methods in terms of correct early warning quantity and false alarm quantity.
The invention provides an early-stage icing early-warning method for a power transmission line based on a fiber vibration data level context, which can adapt to differences between vibration data of thinner icing and non-icing in the early stage of the power transmission line, and early-stage icing early-warning for the power transmission line due to different power transmission line construction.
Specifically, the invention constructs a fiber vibration data level context characteristic extraction module and a hierarchical icing decision neural network model, and utilizes the characteristics of fiber vibration data established hierarchically and finds the difference between the vibration data of thinner icing and non-icing at the initial stage of the transmission line, thereby realizing early-warning of icing at the initial stage of the transmission line.
The invention can describe the change process between different periods of the optical fiber vibration data through the hierarchical structure, on one hand, the neural network can find out the corresponding icing early warning of a specific level, adapt to the characteristic change of the optical fiber vibration data caused by different construction modes and assembly modes, on the other hand, the difference calculated by a formula between the levels can be independent of samples, the difference between different periods of the optical fiber vibration data can be expressed more clearly, and the invention brings help for the situation of fewer initial icing samples of the transmission line. The method and the system can more effectively perform early-stage icing early warning of the power transmission line and improve the efficiency of power grid inspection management.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which is also intended to be covered by the present invention.
Claims (7)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411787705.2A CN119249136B (en) | 2024-12-06 | 2024-12-06 | Early-stage icing early-warning method for power transmission line based on optical fiber vibration data level context |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411787705.2A CN119249136B (en) | 2024-12-06 | 2024-12-06 | Early-stage icing early-warning method for power transmission line based on optical fiber vibration data level context |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119249136A true CN119249136A (en) | 2025-01-03 |
| CN119249136B CN119249136B (en) | 2025-02-11 |
Family
ID=94021183
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411787705.2A Active CN119249136B (en) | 2024-12-06 | 2024-12-06 | Early-stage icing early-warning method for power transmission line based on optical fiber vibration data level context |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119249136B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103616101A (en) * | 2013-11-25 | 2014-03-05 | 国家电网公司 | Method for detecting optical fiber composite ground wire icing state of electric transmission line |
| CN110686626A (en) * | 2019-10-14 | 2020-01-14 | 南京大学 | An OPGW icing monitoring system and method based on a phase sensitive optical time domain reflectometry system |
| US12123995B1 (en) * | 2023-05-16 | 2024-10-22 | China University Of Mining And Technology | Intelligent monitoring and early warning device and method for rock burst based on multi-field and multi-source information fusion |
-
2024
- 2024-12-06 CN CN202411787705.2A patent/CN119249136B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103616101A (en) * | 2013-11-25 | 2014-03-05 | 国家电网公司 | Method for detecting optical fiber composite ground wire icing state of electric transmission line |
| CN110686626A (en) * | 2019-10-14 | 2020-01-14 | 南京大学 | An OPGW icing monitoring system and method based on a phase sensitive optical time domain reflectometry system |
| US12123995B1 (en) * | 2023-05-16 | 2024-10-22 | China University Of Mining And Technology | Intelligent monitoring and early warning device and method for rock burst based on multi-field and multi-source information fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119249136B (en) | 2025-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108225906B (en) | Corrosion monitoring and identification and fatigue life evaluation method of cable based on computer vision | |
| CN103076394B (en) | Safety evaluation method for ocean platform based on integration of vibration identification frequencies and vibration mode | |
| CN116432475B (en) | Multi-factor coupling collaborative early warning method and system for steel structure fatigue crack growth | |
| CN117128448A (en) | An intelligent monitoring system for LNG storage tank structure | |
| CN118586292B (en) | Life estimation method of cable of cable-stayed bridge based on big data | |
| CN113065223A (en) | A multi-level probability correction method for tower-mast cluster digital twin model | |
| CN120705723B (en) | A method, system and medium for monitoring safety risks of large-scale high-altitude steel structure corridors | |
| CN119885394A (en) | Method for constructing Unity3D bridge digital twin platform integrating Midas-Civil calculation function | |
| CN118350090B (en) | A method, medium and system for evaluating fatigue of steel beam air duct hanger buckle | |
| CN114487643A (en) | On-spot handing-over of extra-high voltage GIL equipment is accepted and is synthesized test platform | |
| CN116384189A (en) | A system and method for state assessment of highway bridges | |
| CN119245603A (en) | Inclinometer safety monitoring method based on multi-parameter fusion | |
| Lei et al. | Fiber Bragg grating smart material and structural health monitoring system based on digital twin drive | |
| CN116226974B (en) | Damage identification method and system based on data fusion and adaptive sparse regularization | |
| Crognale et al. | Fatigue Damage Identification by a Global‐Local Integrated Procedure for Truss‐Like Steel Bridges | |
| CN108549847B (en) | A crack damage identification method for beam structures without datum data | |
| Xu et al. | A method for suspenders tension identification of bridges based on the spatio‐temporal correlation between the girder strain and suspenders tension | |
| CN117807861A (en) | Correlation method of ice-covered microclimate data of power transmission line and meteorological station data | |
| CN119249136B (en) | Early-stage icing early-warning method for power transmission line based on optical fiber vibration data level context | |
| CN113987776B (en) | Construction method and test method of dry-type reactor aging resistance test model | |
| CN119814135B (en) | Optical cable on-line monitoring method and system based on time sequence big data analysis | |
| CN119887181A (en) | Bridge inhaul cable disease detection method and system based on image recognition | |
| CN119538541A (en) | Submarine cable life prediction method based on cloud model algorithm and GM (1,1) | |
| He et al. | Location prediction model of zero value insulator based on PNN | |
| CN118095550A (en) | Earth-rock dam deformation prediction method and device based on correlation mapping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |