CN115389881A - Insulation state evaluation method and device for cable intermediate joint - Google Patents
Insulation state evaluation method and device for cable intermediate joint Download PDFInfo
- Publication number
- CN115389881A CN115389881A CN202211019668.1A CN202211019668A CN115389881A CN 115389881 A CN115389881 A CN 115389881A CN 202211019668 A CN202211019668 A CN 202211019668A CN 115389881 A CN115389881 A CN 115389881A
- Authority
- CN
- China
- Prior art keywords
- basis function
- function model
- radial basis
- partial discharge
- data
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1263—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
- G01R31/1272—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation of cable, line or wire insulation, e.g. using partial discharge measurements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Testing Relating To Insulation (AREA)
Abstract
The invention discloses a method and a device for evaluating the insulation state of a cable intermediate joint, wherein the method comprises the following steps: the method comprises the steps of obtaining a partial discharge characteristic quantity of a cable middle joint and a change trend index related to the combination of a discharge repetition rate and the partial discharge characteristic quantity, establishing a radial basis function model according to the partial discharge characteristic quantity and the change trend index, training and verifying the trained radial basis function model to obtain a target radial basis function model, inputting preprocessed to-be-detected data of the cable middle joint to be detected into the target radial basis function model to obtain state evaluation result data of each sub-network, fusing the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain insulation state result data. The method is favorable for solving the technical problem that the evaluation accuracy is low due to the fact that the state change trend of the cable joint is difficult to evaluate in the data acquisition interval period in the existing insulation state evaluation method, and the timeliness and the accuracy of the insulation state evaluation of the cable middle joint are improved.
Description
Technical Field
The invention relates to the technical field of insulation states of cable intermediate joints, in particular to a method and a device for evaluating the insulation state of a cable intermediate joint.
Background
The cable intermediate joint is mainly used for connecting a cable body and is a key component for ensuring safe and reliable operation of a cable line. With the wide use of a large number of cables in urban power transmission lines, the number of corresponding cable joints is also becoming larger. However, the cable intermediate joint is easily subjected to the influence of factors such as manufacturing level, insulating materials and external environment, and insulation breakdown faults are easily generated, and the cable intermediate joint becomes the biggest weak link of a cable line at present. Long-term operating experience has shown that the progressive deterioration of the insulation of the cable intermediate joint until breakdown is often accompanied by significant partial discharges. Analysis of the related documents reveals that partial discharge is closely related to the insulation condition of the joint, and a change in the amount of partial discharge indicates a deterioration in the insulation of the joint. Therefore, the partial discharge parameter of the cable middle joint can be used as one of important judgment bases for the insulation state evaluation of the cable joint.
At present, the insulation state evaluation of the cable middle joint is mainly based on items such as discharge amount level and insulation resistance detection, the data acquisition interval time is long, the state change trend of the cable joint during the interval is difficult to evaluate, and therefore the real state of equipment cannot be completely reflected. In recent years, the continuous development of online monitoring technology provides a good data source for trend analysis of state variables. How to organically combine the size and the variation trend of the online monitoring data and use the online monitoring data as a judgment basis for cable intermediate joint insulation state evaluation needs to construct a reasonable and effective cable joint insulation state evaluation model with multi-information fusion.
Therefore, in order to improve the timeliness and accuracy of the insulation state evaluation of the cable intermediate joint and solve the technical problem that the evaluation accuracy is low due to the fact that the existing insulation state evaluation method is difficult to evaluate the state change trend of the cable joint during the data acquisition interval, it is urgently needed to construct the insulation state evaluation method of the cable intermediate joint.
Disclosure of Invention
The invention provides an insulation state evaluation method and device for a cable intermediate joint, and solves the technical problem that the existing insulation state evaluation method is low in evaluation accuracy due to the fact that the state change trend of the cable joint is difficult to evaluate in the data acquisition interval.
In a first aspect, the present invention provides a method for evaluating an insulation state of a cable intermediate joint, including:
acquiring a partial discharge characteristic quantity of a cable intermediate joint and a change trend index related to the combination of a discharge repetition rate and the partial discharge characteristic quantity;
establishing a radial basis function model according to the partial discharge characteristic quantity and the change trend index;
training and verifying the trained radial basis function model based on the partial discharge characteristic quantity and the change trend index to obtain a target radial basis function model;
acquiring preprocessed to-be-detected data of a cable intermediate joint to be detected;
inputting the preprocessed to-be-detected data into the target radial basis function model to obtain state evaluation result data of each sub-network of the target radial basis function model;
and fusing the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain the insulation state result data of the intermediate joint of the cable to be tested.
Optionally, the obtaining of the characteristic quantity of partial discharge of the cable intermediate joint and the index of variation trend regarding the repetition rate of discharge combined with the characteristic quantity of partial discharge comprises:
acquiring original partial discharge signal data of a cable intermediate joint and performing feature extraction on the original partial discharge signal to obtain original partial discharge feature quantity;
according to preset relative deterioration degree data, carrying out standardization processing on the original partial discharge characteristic quantity to obtain the partial discharge characteristic quantity of the cable intermediate joint;
and constructing a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity according to a preset valence trend parameter and the partial discharge characteristic quantity.
Optionally, training and verifying the trained radial basis function model based on the partial discharge characteristic quantity and the variation trend index to obtain a target radial basis function model, including:
dividing the partial discharge characteristic quantity and the change trend index into training data and verification data;
training the radial basis function model according to the training data to obtain a trained radial basis function model;
and verifying the trained radial basis function model based on the verification data to obtain the target radial basis function model.
Optionally, training the radial basis function model according to the training data to obtain a trained radial basis function model, including:
inputting the training data into the radial basis function model to obtain corresponding insulation state prediction result data;
determining a training error according to a data label corresponding to the training data and the insulation state prediction result data;
and adjusting the radial basis function model based on the training error to obtain an optimal parameter, and optimizing the radial basis function model by adopting the optimal parameter to obtain the trained radial basis function model.
Optionally, before the training data is input into the radial basis function model and corresponding insulation state prediction result data is obtained, the method further includes:
parameters of the radial basis function model are initialized.
In a second aspect, the present invention provides an insulation state evaluation device for a cable intermediate joint, comprising:
the acquisition module is used for acquiring the partial discharge characteristic quantity of the cable intermediate joint and a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity;
the establishing module is used for establishing a radial basis function model according to the partial discharge characteristic quantity and the change trend index;
the training module is used for training and verifying the trained radial basis function model based on the partial discharge characteristic quantity and the change trend index to obtain a target radial basis function model;
the module to be tested is used for acquiring the preprocessed data to be tested of the intermediate joint of the cable to be tested;
the evaluation module is used for inputting the preprocessed data to be tested into the target radial basis function model to obtain state evaluation result data of each sub-network of the target radial basis function model;
and the fusion module is used for fusing the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain the insulation state result data of the intermediate joint of the cable to be tested.
Optionally, the obtaining module includes:
the acquisition submodule is used for acquiring original partial discharge signal data of a cable intermediate connector and extracting the characteristics of the original partial discharge signal to obtain original partial discharge characteristic quantity;
the processing submodule is used for carrying out standardization processing on the original partial discharge characteristic quantity according to preset relative deterioration degree data to obtain the partial discharge characteristic quantity of the cable intermediate joint;
and the construction submodule is used for constructing a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity according to a preset valence trend parameter and the partial discharge characteristic quantity.
Optionally, the training module comprises:
the dividing submodule is used for dividing the partial discharge characteristic quantity and the change trend index into training data and verification data;
the training submodule is used for training the radial basis function model according to the training data to obtain a trained radial basis function model;
and the verification sub-module is used for verifying the trained radial basis function model based on the verification data to obtain the target radial basis function model.
Optionally, the training submodule includes:
the training unit is used for inputting the training data into the radial basis function model to obtain corresponding insulation state prediction result data;
the determining unit is used for determining a training error according to the data label corresponding to the training data and the insulation state prediction result data;
and the optimization unit is used for adjusting the radial basis function model based on the training error to obtain an optimal parameter, and optimizing the radial basis function model by adopting the optimal parameter to obtain the trained radial basis function model.
Optionally, the training sub-module further comprises:
an initialization unit for initializing parameters of the radial basis function model.
According to the technical scheme, the invention has the following advantages: the invention provides an insulation state evaluation method of a cable intermediate joint, which solves the technical problem that the existing insulation state evaluation method has low evaluation accuracy due to the fact that the state change trend of the cable intermediate joint is difficult to evaluate during the data acquisition interval by applying a D-S evidence theory and combining with a preset decision rule, and obtains the insulation state result data of the cable intermediate joint to be measured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
FIG. 1 is a flowchart illustrating a first embodiment of a method for evaluating an insulation state of a cable intermediate joint according to the present invention;
FIG. 2 is a flowchart illustrating a second embodiment of a method for evaluating an insulation state of a cable intermediate joint according to the present invention;
FIG. 3 is a diagram of a trend indicator relating to the combination of the discharge repetition rate and the partial discharge characteristic according to the present invention;
FIG. 4 is a diagram of a trend indicator relating to the combination of the discharge repetition rate and the partial discharge characteristic according to the present invention;
fig. 5 is a block diagram illustrating an embodiment of an insulation state evaluating apparatus for a cable intermediate joint according to the present invention.
Detailed Description
The embodiment of the invention provides an insulation state evaluation method and device for a cable intermediate joint, which are used for solving the technical problem of low evaluation accuracy caused by difficulty in evaluating the state change trend of the cable joint during a data acquisition interval in the existing insulation state evaluation method.
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart illustrating a first method for evaluating an insulation state of a cable intermediate connector according to a first embodiment of the present invention, including:
step S101, acquiring partial discharge characteristic quantity of a cable intermediate joint and a change trend index related to combination of a discharge repetition rate and the partial discharge characteristic quantity;
in the embodiment of the invention, the method comprises the steps of obtaining original partial discharge signal data of a cable intermediate joint and carrying out feature extraction on the original partial discharge signal to obtain original partial discharge feature quantity, carrying out standardization processing on the original partial discharge feature quantity according to preset relative degradation degree data to obtain the partial discharge feature quantity of the cable intermediate joint, and constructing a change trend index related to the combination of a discharge repetition rate and the partial discharge feature quantity according to a preset valence trend parameter and the partial discharge feature quantity.
Step S102, establishing a radial basis function model according to the partial discharge characteristic quantity and the change trend index;
step S103, training and verifying the trained radial basis function model based on the partial discharge characteristic quantity and the change trend index to obtain a target radial basis function model;
in the embodiment of the present invention, the partial discharge characteristic quantity and the change trend index are divided into training data and verification data, the radial basis function model is trained according to the training data to obtain a trained radial basis function model, and the trained radial basis function model is verified based on the verification data to obtain the target radial basis function model.
Step S104, data to be tested of the cable intermediate joint to be tested;
step S105, inputting the preprocessed data to be tested into the target radial basis function model to obtain state evaluation result data of each sub-network of the target radial basis function model;
and S106, fusing the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain insulation state result data of the intermediate joint of the cable to be tested.
According to the insulation state evaluation method of the cable intermediate joint provided by the embodiment of the invention, the problem of low evaluation accuracy of the existing insulation state evaluation method due to the fact that the state change trend of the cable intermediate joint is difficult to evaluate during the data acquisition interval is solved and the insulation state evaluation accuracy of the cable intermediate joint is improved through the insulation state evaluation method of the cable intermediate joint.
In a second embodiment, referring to fig. 2, fig. 2 is a flowchart illustrating a method for evaluating an insulation state of a cable intermediate connector according to the present invention, including:
step S201, acquiring original partial discharge signal data of a cable intermediate joint and performing feature extraction on the original partial discharge signal to obtain original partial discharge feature quantity;
in the embodiment of the invention, the original partial discharge signal data of the cable intermediate joint is obtained and the characteristic extraction is carried out on the original partial discharge signal to obtain the original partial discharge characteristic quantity, including the discharge repetition rate n and the maximum discharge quantity Q m Average discharge amount Q av Partial discharge energy maximum W m Pulse phase width
In the specific implementation, a high-frequency current sensor is used for obtaining a partial discharge signal, and the partial discharge signal is subjected to filtering denoising, attenuation amplification, A/D conversion, feature extraction and other steps to obtain a partial discharge feature set, wherein the selected feature set comprises a discharge repetition rate n and a maximum discharge Q m Average discharge amount Q av Partial discharge energy maximum W m Pulse phase width
The discharge repetition rate n refers to the number of partial discharges in unit time, and the more severe the partial discharge phenomenon and the higher the discharge repetition rate with the continuous degradation of the middle joint insulation, the calculation formula is specifically as follows:
wherein M represents the total number of detected power frequency cycles, N s Representing the discharge frequency of the s-th detection power frequency period.
Maximum discharge Q m Refers to the single maximum discharge during the detection period. In the process of the insulation deterioration degree of the cable middle joint or the process of the breakdown, the sudden change phenomenon that the local discharge amount is greatly increased is often accompanied. Thus, the value visually reflects the severity of the insulation damage to the joint.
Average discharge capacity Q av The calculation formula is specifically shown as a discharge repetition rate calculation formula, and is an arithmetic mean value of the partial discharge quantities of all the partial discharge signals detected in the detection period. This value can be used as a reference for the overall intensity of the partial discharge signal.
Pulse phase width
Refers to the width of the phase interval where the discharge pulse is after removing the background noise, wherein the widths of the positive and negative half-cycle discharge pulses in one power frequency cycle are respectively called
As the partial discharge continues, the pulse phase width of the positive and negative half cycles increases accordingly, and therefore the degree of development of the partial discharge is reflected to a certain extent.
Maximum value of partial discharge energy W m Means the local discharge energy value W of all partial discharge signals in the detection period i Maximum value of (1), wherein W i The energy of the ith discharge pulse in the detection period is detected, and the calculation formula is shown as (2). Since the destruction of the PD to the insulation is necessarily accompanied by the exchange of energy, the discharge energy and the insulation destruction are closely related, and generally, the more the insulation deterioration, the larger the discharge energy.
In addition, the average discharge amount is determined by the following equation:
wherein n represents the number of discharges, Q i Indicating the apparent discharge amount detected by the i-th discharge,
wherein q is i Represents the apparent discharge amount, mu, detected by the i-th discharge i Indicating the initial discharge voltage of the i-th discharge.
Step S202, according to preset relative deterioration degree data, carrying out standardization processing on the original partial discharge characteristic quantity to obtain the partial discharge characteristic quantity of the cable intermediate joint;
in the embodiment of the invention, relative degradation degree data is introduced, and the original partial discharge characteristic quantity is subjected to standardization processing to obtain the partial discharge characteristic quantity of the cable intermediate joint.
In a specific implementation, the relative degradation degree is introduced to carry out standardization processing on each characteristic quantity.
The deterioration degree refers to the degree of deterioration compared with the actual operation state and the fault state of the current equipment, and the value range is [0,1]. Different degradation degree values represent that the equipment is in different running states, and the mapping relation between the value range of the degradation degree of the intermediate joint and the running states is shown in the following table:
the smaller and more optimal indexes are adopted for standardization treatment, and the formula is specifically as follows:
wherein x is min 、x max Respectively, a lower threshold and an upper threshold for evaluating the characteristic quantity, the values of which are often determined according to a preventive test protocol or a handover test protocol. The value range is shown in the following table (single cycle, in one power frequency cycle):
step S203, constructing a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity according to a preset valence trend parameter and the partial discharge characteristic quantity;
in the embodiment of the invention, a valence trend parameter is introduced, and a variation trend index which is related to the combination of the discharge repetition rate and the partial discharge characteristic quantity is constructed.
In a particular implementation, the degradation of the insulation of the cable intermediate joint is not a sudden process, but a process that evolves progressively. In the process, each characteristic quantity of the partial discharge shows a stepwise change trend along with the accumulation of time and has certain fluctuation. Therefore, the on-line monitoring data does not completely reflect the insulation state of the cable joint. For example, in actual engineering, if a certain index is at a warning value but always keeps a steady development trend, the cable joint is determined to still work normally.
However, the uncertainty of the partial discharge causes the development trend of each partial discharge characteristic quantity to be difficult to quantify. By drawing a scatter diagram between two different characteristic quantities, strong correlation between the partial discharge characteristic quantities can be judged. Referring to fig. 3, fig. 3 is a schematic diagram of a variation trend index related to a combination of a discharge repetition rate and a partial discharge characteristic amount according to the present invention, and fig. 4 is a schematic diagram of a variation trend index related to a combination of a discharge repetition rate and a partial discharge characteristic amount according to the present invention, wherein an X-axis is a discharge repetition rate and a Y-axis is a partial discharge characteristic amount. From the distribution of the data points, the independent variable x and the dependent variable y have basically the same change trend; then, the linear correlation degrees of the two parameters are respectively obtained by applying the Pearson correlation coefficient, when the correlation coefficient is larger than 0.75, the correlation of the two parameters is considered to be high, otherwise, the correlation is considered to be a redundant parameter and eliminated, and the calculation formula is specifically as follows:
in the above analysis, r is a correlation sum, and the discharge repetition rate has a strong linear correlation with other partial discharge characteristic quantities. In order to describe the variation trend of the discharge repetition rate and the rest partial discharge characteristic quantity, the invention introduces a valence trend (PVT) parameter into partial discharge trend analysis to obtain a trend index of the partial discharge characteristic quantity-discharge repetition rate. Price volume trends, i.e., price-volume trends, are used to reveal trends in stock changes that are difficult to describe for single transaction prices and volumes.
The PVT parameter S is obtained by the following method for any partial discharge characteristic quantity P PVT The simplified formula of (c):
wherein n is a discharge repetition rate; t is time; delta S PVT Increment of PVT parameters for 2 successive units of calculation, definition of Δ P and Δ S PVT Similarly.
And (4) solving a PVT (pressure-volume-temperature) graph of the partial discharge characteristic quantity-discharge repetition rate of the cable joint according to the formula. In order to quantify the change trend of each partial discharge characteristic quantity, the invention firstly equally divides the trend graph of each partial discharge characteristic quantity into 6 parts according to the total data length, then calculates 1-order and 2-order derivative parameters at 5 equally divided points as the quantitative value basis of the 1-order (2-order) derivative of the corresponding statistical parameter, and finally carries out specific value taking according to the following rules: when the derivatives of 1 st order and 2 nd order at 5 bisectors of a trend chart of a certain partial discharge characteristic quantity are both positive (negative), the result value is taken as '1' ('-2'); when both positive (negative) and 0 values are present, the resulting value takes the value "2" ("1"); when there is both a positive value and a negative value, the resulting value is in an indeterminate state and is taken as "0".
Step S204, establishing a radial basis function model according to the partial discharge characteristic quantity and the change trend index;
in the embodiment of the invention, a radial basis function model is established according to the partial discharge characteristic quantity and the change trend index.
In the concrete implementation, on-line monitoring data of average discharge capacity, partial discharge capacity, positive and negative phase pulse width and maximum discharge energy are selected as parameter subspace I 1 And is combined with 1 The variation trend of each parameter in the space is taken as an independent parameter subspace I 2 . According to the existing insulation state evaluation method based on the neural network, the invention adopts the Radial Base Function (RBF) neural network with the best approximation performance as the first-layer evaluation subsystem, wherein the included artificial neural networks respectively correspond to the parameter subspace I 1 And I 2 In turn denoted as RBF 1 ，RBF 2 . In addition, as can be seen from the table, the output of the neural network should include 4 nodes, i.e., state subspace S = { S = } 1 ，S 2 ，S 3 ，S 4 }，S 1 ～S 4 Corresponding to 4 states of normal, caution, abnormal and severe, respectively.
Step S205, dividing the partial discharge characteristic quantity and the change trend index into training data and verification data;
step S206, training the radial basis function model according to the training data to obtain a trained radial basis function model;
in an optional embodiment, training the radial basis function model according to the training data to obtain a trained radial basis function model, includes:
initializing parameters of the radial basis function model;
inputting the training data into the radial basis function model to obtain corresponding insulation state prediction result data;
determining a training error according to a data label corresponding to the training data and the insulation state prediction result data;
and adjusting the radial basis function model based on the training error to obtain an optimal parameter, and optimizing the radial basis function model by adopting the optimal parameter to obtain the trained radial basis function model.
In the embodiment of the invention, the parameters of the radial basis function model are initialized, the training data are input into the radial basis function model to obtain the corresponding insulation state prediction result data, the training error is determined according to the data label corresponding to the training data and the insulation state prediction result data, the radial basis function model is adjusted based on the training error to obtain the optimal parameters, and the radial basis function model is optimized by adopting the optimal parameters to obtain the trained radial basis function model.
In the concrete implementation, in the primary radial basis function model training process, a plurality of groups of on-line monitoring data are collected as training samples for the constructed RBF 1 、RBF 2 Training a neural network, wherein the number of 2 sub-network input layer nodes is the number of state quantities in the network, and the number is 6 and 5 respectively; the number of the output layer nodes is 4, and the output layer nodes correspond to 4 states respectively. Since the magnitude of the input data of the neural network is usually greatly different, if the raw data is directly input, the network is not sensitive to smaller values, and some important features are difficult to acquire by the network. So the relative degradation degree pretreatment is adopted. The specific training process of the RBF network is as follows:
1) And initializing RBF network parameters, and giving values of eta and alpha and iteration termination precision epsilon.
Determining an input vector X = [ X ] 1 ,x 2 ,...,x n ] T Wherein n is the number of neurons in the input layer.
Determining an output vector Y = [ Y ] 1 ,y 2 ,...,y q ] T And desired output O = [ O ] 1 ,o 2 ,...o q ] T And q is the number of output neurons.
Initializing connection weight W from hidden layer to output layer k ＝[w k1 ,w k2 ,...w kp ] T Wherein k = (1, 2.. Q),
wherein p is the number of cryptic neurons; q is the number of neurons in the output layer k min Is the minimum of all expected outputs in the kth output neuron in the training set; k is a radical of max Is in the training set; the maximum of all expected outputs in the kth output neuron.
Initializing central parameters of each neuron of the hidden layer;
the central vector of each neuron of the hidden layer is C j ＝[c j1 ,c j2 ,...c jn ] T (ii) a Wherein, the central parameters are specifically as follows:
wherein, c ji Is the central parameter of the central vector of the ith feature of the jth neuron of the hidden layer, p is the total number of neurons of the hidden layer, i min Is the minimum value of all input information of the ith feature in the training set, i max There is a maximum value of the input information for the ith feature in the training set.
Initializing a width vector;
width vector of
The width adjusting coefficient is specifically as follows:
wherein d is f The width adjustment coefficient has a value less than 1, and has the effect of enabling each hidden layer neuron to easily realize the sensing capability on local information, thereby being beneficial to improving the local response capability of the RBF neural network.
2) Calculating the output value of the jth neuron of the hidden layer, wherein the calculation formula specifically comprises the following steps:
wherein z is j The output value of the jth neuron of the hidden layer is shown, and X is an input vector; c j Is the central vector of the jth neuron of the hidden layer; d j Is the width vector of the jth neuron; | | | is the euclidean norm.
3) Computing output of output layer neurons Y = [ Y 1 ,y 2 ,...y k ,...y q ] T Element y thereof k Is shown as
Wherein, w kj Is the adjustment weight between the kth neuron of the output layer and the jth neuron of the hidden layer.
4) The loss function for calculating the network output is specifically as follows:
5) Iterative computation of weight parameters
If RMS ≦ ε or N equals the maximum number of neurons N m And if so, stopping the network training. Otherwise, retraining parameters such as RBF neural network weight, center, width and the like by using a gradient descent method, wherein the iterative calculation formula is specifically as follows:
wherein, w kj (t) the adjustment weight between the kth output neuron and the jth hidden layer neuron at the time of the t iteration calculation; c. C ji (t) a central component of the jth hidden layer neuron for the ith input neuron at the time of the tth iterative computation; d ji (t) is the width corresponding to the center; eta is the learning rate; e is an RBF neural network evaluation function, and specifically comprises the following steps:
wherein o is lk Is the expected output value of the kth output neuron at the ith input sample; y is lk The net output value of the kth output neuron at the ith input sample is obtained.
Step S207, verifying the trained radial basis function model based on the verification data to obtain the target radial basis function model;
step S208, acquiring preprocessed to-be-detected data of the intermediate joint of the to-be-detected cable;
in the embodiment of the invention, the preprocessed data to be detected of the intermediate joint of the cable to be detected is obtained, and the preprocessed data to be detected is the partial discharge characteristic quantity of the intermediate joint of the cable to be detected and the change trend index of the combination of the discharge repetition rate and the partial discharge characteristic quantity.
Step S209, inputting the preprocessed data to be detected into the target radial basis function model to obtain state evaluation result data of each sub-network of the target radial basis function model;
in the embodiment of the invention, the preprocessed data to be detected is input into the target radial basis function model, and the state evaluation result data of each sub-network of the target radial basis function model is obtained through calculation.
Step S210, fusing the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain insulation state result data of the intermediate joint of the cable to be tested;
in the embodiment of the invention, according to a D-S evidence theory and a preset decision rule, the state evaluation result data is fused to obtain the insulation state result data of the intermediate joint of the cable to be tested so as to evaluate the insulation state of the intermediate joint of the cable to be tested.
In the specific implementation, the insulation state of the intermediate joint of the cable to be tested is evaluated based on the fusion of the D-S evidence theory.
1) Constructing an identification frame: Θ = { S 1 ,S 2 ,S 3 ,S 4 And the states correspond to normal, attention, abnormal and severe states respectively.
2) And taking the state evaluation result data of each sub-network as an independent evidence body and an independent evidence set.
3) Determining a basic probability distribution function for each evidence: the output values of the diagnosis sub-networks are converted to be used as basic probability distribution for proposition on the identification framework, and the complexity of building a basic probability distribution function is avoided while the objectification of the basic probability distribution assignment is realized. Since the diagnostic capabilities and scope of each diagnostic subnetwork are different, there is a discount on the reliability factor, i.e., evidence, for each network. Constructing the following confusion matrix according to the training result of each RBF subnetwork, and calculating by a formula to obtain the accuracy, namely the confidence of some evidence body, specifically:
let the jth output value of the ith sub-network be O i (j) Then, the basic probability distribution of the network to the judgment j is specifically as follows:
m i (Θ)＝1-α i (i＝1,2,...,p)；
wherein m is i (j) Representing the trust distribution of the ith evidence to the judgment j; alpha is alpha i Representing the degree of trust of the ith evidence body; m is i (Θ) represents the uncertain trust assignment for the ith evidence body; q represents the number of output values of the ith network; p represents the number of evidential bodies.
4) And calculating the reliability function and the plausibility function of each state type by using the determined basic probability distribution.
5) Evidence combination rules of D-S evidence theory:
wherein the content of the first and second substances,
wherein m is 1 (X) and m 2 (Y) is the basic confidence function 1 for X (evidence body 1) and the basic confidence function 2 for Y (evidence body 2), respectively. The value of K indicates the degree to which the combined evidence conflicts with each other. When K =0, 2 evidences are completely consistent (completely consistent); when K =1,2 evidences are completely conflicting; when 0 is present<K<1, indicates that the 2 evidence portions are consistent.
6) Evaluating the decision to obtain a confidence interval [ Bel ] of the evidence to all the state types in the recognition frame j ,pl j ]And uncertainty m of evidence i After (Θ), the cable joint insulation state evaluation result can be determined by the following rule.
Is provided with S 1 ,
Satisfies the following conditions:
wherein S is 1 、S 2 To identify any of the different types of states within the framework; m (S) 1 )，m(S 2 ) Respectively representing a maximum and a second maximum confidence function value; and m (theta) is an uncertain reliability function value.
It is noted that the threshold ε 1 The value of the cable joint needs to be combined with the actual cable joint insulation condition and the requirement of effectively distinguishing the joint insulation state, and the value is more suitable between 0.4 and 0.5 under the general condition; in addition, the threshold ε 2 Must be greater than the uncertain confidence function values of each evidence body, so that epsilon can be determined by selecting the reliability coefficient of each evaluation network 2 The value of (c).
If S 1 Satisfy m i (ΘΘ＝1-α i (i =1,2,. Gtang, p), then S 1 To evaluate the results. If the above rules cannot be satisfied simultaneously, the evaluation result cannot be obtained. This may occur for two reasons: (1) The state type is not in the current recognition frame, so the recognition frame must be determined again; (2) The evidence selection is not reasonable, so that more evidences need to be selected again or further selected for fusion calculation.
According to the insulation state evaluation method of the cable intermediate joint provided by the embodiment of the invention, the problem of low evaluation accuracy of the existing insulation state evaluation method due to the fact that the state change trend of the cable intermediate joint is difficult to evaluate during the data acquisition interval is solved and the insulation state evaluation accuracy of the cable intermediate joint is improved through the insulation state evaluation method of the cable intermediate joint.
Referring to fig. 5, fig. 5 is a block diagram of an insulation state evaluation apparatus for a cable intermediate connector according to an embodiment of the present invention, including:
an obtaining module 501, configured to obtain a partial discharge characteristic quantity of a cable intermediate joint and a variation trend index related to a combination of a discharge repetition rate and the partial discharge characteristic quantity;
an establishing module 502, configured to establish a radial basis function model according to the partial discharge characteristic quantity and the change trend indicator;
a training module 503, configured to train and verify the trained radial basis function model based on the partial discharge characteristic quantity and the change trend index, to obtain a target radial basis function model;
a module to be tested 504, configured to obtain pre-processed data to be tested of the intermediate joint of the cable to be tested;
an evaluation module 505, configured to input the preprocessed data to be tested into the target radial basis function model, to obtain state evaluation result data of each sub-network of the target radial basis function model;
and the fusion module 506 is configured to fuse the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain insulation state result data of the intermediate joint of the cable to be tested.
Optionally, the obtaining module 501 includes:
the acquisition submodule is used for acquiring original partial discharge signal data of a cable intermediate connector and extracting the characteristics of the original partial discharge signal to obtain original partial discharge characteristic quantity;
the processing submodule is used for carrying out standardization processing on the original partial discharge characteristic quantity according to preset relative deterioration degree data to obtain the partial discharge characteristic quantity of the cable intermediate joint;
and the construction submodule is used for constructing a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity according to a preset valence trend parameter and the partial discharge characteristic quantity.
Optionally, the training module 503 comprises:
the dividing submodule is used for dividing the partial discharge characteristic quantity and the change trend index into training data and verification data;
the training submodule is used for training the radial basis function model according to the training data to obtain a trained radial basis function model;
and the verification sub-module is used for verifying the trained radial basis function model based on the verification data to obtain the target radial basis function model.
Optionally, the training submodule includes:
the training unit is used for inputting the training data into the radial basis function model to obtain corresponding insulation state prediction result data;
the determining unit is used for determining a training error according to the data label corresponding to the training data and the insulation state prediction result data;
and the optimization unit is used for adjusting the radial basis function model based on the training error to obtain an optimal parameter, and optimizing the radial basis function model by adopting the optimal parameter to obtain the trained radial basis function model.
Optionally, the training sub-module further comprises:
an initialization unit for initializing parameters of the radial basis function model.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present application, it should be understood that the method and apparatus disclosed in the present invention can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a readable storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned readable storage medium comprises: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for evaluating an insulation state of a cable intermediate joint, comprising:
acquiring a partial discharge characteristic quantity of a cable intermediate joint and a change trend index related to the combination of a discharge repetition rate and the partial discharge characteristic quantity;
establishing a radial basis function model according to the partial discharge characteristic quantity and the change trend index;
training and verifying the trained radial basis function model based on the partial discharge characteristic quantity and the change trend index to obtain a target radial basis function model;
acquiring preprocessed to-be-detected data of a cable intermediate joint to be detected;
inputting the preprocessed to-be-detected data into the target radial basis function model to obtain state evaluation result data of each sub-network of the target radial basis function model;
and fusing the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain the insulation state result data of the intermediate joint of the cable to be tested.
2. The insulation state evaluation method of a cable intermediate joint according to claim 1, wherein obtaining a partial discharge characteristic quantity of the cable intermediate joint and a variation tendency index regarding a discharge repetition rate in combination with the partial discharge characteristic quantity includes:
acquiring original partial discharge signal data of a cable intermediate joint and performing feature extraction on the original partial discharge signal to obtain original partial discharge feature quantity;
according to preset relative deterioration degree data, carrying out standardization processing on the original partial discharge characteristic quantity to obtain the partial discharge characteristic quantity of the cable intermediate joint;
and constructing a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity according to a preset valence trend parameter and the partial discharge characteristic quantity.
3. The insulation state evaluation method of a cable intermediate joint according to claim 1, wherein training and verifying the trained radial basis function model based on the partial discharge characteristic quantity and the variation tendency index to obtain a target radial basis function model comprises:
dividing the partial discharge characteristic quantity and the change trend index into training data and verification data;
training the radial basis function model according to the training data to obtain a trained radial basis function model;
and verifying the trained radial basis function model based on the verification data to obtain the target radial basis function model.
4. The method for evaluating the insulation state of the intermediate joint of the cable according to claim 3, wherein training the radial basis function model according to the training data to obtain a trained radial basis function model comprises:
inputting the training data into the radial basis function model to obtain corresponding insulation state prediction result data;
determining a training error according to a data label corresponding to the training data and the insulation state prediction result data;
and adjusting the radial basis function model based on the training error to obtain an optimal parameter, and optimizing the radial basis function model by adopting the optimal parameter to obtain the trained radial basis function model.
5. The method of claim 4, wherein before inputting the training data into the radial basis function model to obtain corresponding insulation state prediction result data, the method further comprises:
parameters of the radial basis function model are initialized.
6. An insulation state evaluation device of a cable intermediate joint, characterized by comprising:
the acquisition module is used for acquiring the partial discharge characteristic quantity of the cable intermediate joint and a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity;
the establishing module is used for establishing a radial basis function model according to the partial discharge characteristic quantity and the change trend index;
the training module is used for training and verifying the trained radial basis function model based on the partial discharge characteristic quantity and the change trend index to obtain a target radial basis function model;
the module to be tested is used for acquiring preprocessed data to be tested of the intermediate joint of the cable to be tested;
the evaluation module is used for inputting the preprocessed data to be tested into the target radial basis function model to obtain state evaluation result data of each sub-network of the target radial basis function model;
and the fusion module is used for fusing the state evaluation result data by using a D-S evidence theory and combining a preset decision rule to obtain the insulation state result data of the intermediate joint of the cable to be tested.
7. The insulation state evaluation device of a cable intermediate joint according to claim 6, wherein the acquisition module comprises:
the acquisition submodule is used for acquiring original partial discharge signal data of a cable intermediate connector and extracting the characteristics of the original partial discharge signal to obtain original partial discharge characteristic quantity;
the processing submodule is used for carrying out standardization processing on the original partial discharge characteristic quantity according to preset relative deterioration degree data to obtain the partial discharge characteristic quantity of the cable intermediate joint;
and the construction submodule is used for constructing a change trend index related to the combination of the discharge repetition rate and the partial discharge characteristic quantity according to a preset valence trend parameter and the partial discharge characteristic quantity.
8. The insulation state evaluation device of a cable intermediate joint according to claim 6, wherein the training module comprises:
the dividing submodule is used for dividing the partial discharge characteristic quantity and the change trend index into training data and verification data;
the training submodule is used for training the radial basis function model according to the training data to obtain a trained radial basis function model;
and the verification sub-module is used for verifying the trained radial basis function model based on the verification data to obtain the target radial basis function model.
9. The insulation state evaluation device of a cable intermediate joint according to claim 8, wherein the training submodule includes:
the training unit is used for inputting the training data into the radial basis function model to obtain corresponding insulation state prediction result data;
the determining unit is used for determining a training error according to the data label corresponding to the training data and the insulation state prediction result data;
and the optimization unit is used for adjusting the radial basis function model based on the training error to obtain an optimal parameter, and optimizing the radial basis function model by adopting the optimal parameter to obtain the trained radial basis function model.
10. The insulation state evaluation device of a cable intermediate joint according to claim 9, wherein the training submodule further comprises:
an initialization unit for initializing parameters of the radial basis function model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211019668.1A CN115389881A (en) | 2022-08-24 | 2022-08-24 | Insulation state evaluation method and device for cable intermediate joint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211019668.1A CN115389881A (en) | 2022-08-24 | 2022-08-24 | Insulation state evaluation method and device for cable intermediate joint |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115389881A true CN115389881A (en) | 2022-11-25 |
Family
ID=84122217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211019668.1A Pending CN115389881A (en) | 2022-08-24 | 2022-08-24 | Insulation state evaluation method and device for cable intermediate joint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115389881A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117849560A (en) * | 2024-03-07 | 2024-04-09 | 南京中鑫智电科技有限公司 | Valve side sleeve insulation monitoring method and system combining end screen voltage and partial discharge |
-
2022
- 2022-08-24 CN CN202211019668.1A patent/CN115389881A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117849560A (en) * | 2024-03-07 | 2024-04-09 | 南京中鑫智电科技有限公司 | Valve side sleeve insulation monitoring method and system combining end screen voltage and partial discharge |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113702895B (en) | Online quantitative evaluation method for error state of voltage transformer | |
| CN104035431B (en) | The acquisition methods of kernel functional parameter and system for non-linear process monitoring | |
| CN113111591B (en) | Automatic diagnosis method, device and equipment based on internal fault of modular power distribution terminal | |
| CN111967791B (en) | Equipment performance evaluation method and system considering disturbance and fault threshold | |
| CN115389881A (en) | Insulation state evaluation method and device for cable intermediate joint | |
| CN116610998A (en) | Switch cabinet fault diagnosis method and system based on multi-mode data fusion | |
| CN112287556A (en) | Method and device for determining insulation state of cable | |
| CN113657622A (en) | Method, device, terminal and storage medium for fusing multidimensional state data of electrical equipment | |
| CN115600136A (en) | High-voltage bushing fault diagnosis method, system and medium based on multiple sensors | |
| CN114528721A (en) | Cable intermediate joint insulation state assessment method and system | |
| He et al. | A systematic method of remaining useful life estimation based on physics-informed graph neural networks with multisensor data | |
| Wang et al. | Multi‐harmonic sources location based on sparse component analysis and complex independent component analysis | |
| Jia et al. | Partial discharge pattern recognition using variable predictive model‐based class discrimination with kernel partial least squares regression | |
| CN117272102A (en) | Transformer fault diagnosis method based on double-attention mechanism | |
| CN112149791A (en) | Low-voltage risk assessment system based on neural network recognition and fuzzy analysis | |
| Dang et al. | seq2graph: Discovering dynamic non-linear dependencies from multivariate time series | |
| CN116702005A (en) | Neural network-based data anomaly diagnosis method and electronic equipment | |
| Castagliola et al. | Evaluation of the statistical performance of a variable sampling interval R EWMA control chart | |
| Yu et al. | Design for an SPRT control scheme based on linguistic data | |
| CN115423370A (en) | Relay protection equipment health state assessment method and device | |
| CN112069621B (en) | Method for predicting residual service life of rolling bearing based on linear reliability index | |
| Yu et al. | Knowledge reasoning of transmission line component detection using CRITIC and TOPSIS approaches | |
| CN116882333B (en) | Chip limiting frequency prediction method, device, equipment and medium | |
| CN112114215A (en) | Transformer aging evaluation method and system based on error back propagation algorithm | |
| CN113657623B (en) | Power equipment state diagnosis effect determining method, device, terminal and storage medium |
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 |