CN116975674A - Fault optical cable prediction method and device, electronic equipment and storage medium - Google Patents
Fault optical cable prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116975674A CN116975674A CN202210398229.XA CN202210398229A CN116975674A CN 116975674 A CN116975674 A CN 116975674A CN 202210398229 A CN202210398229 A CN 202210398229A CN 116975674 A CN116975674 A CN 116975674A
- Authority
- CN
- China
- Prior art keywords
- optical cable
- parameter
- target
- sample set
- labels
- 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
- 230000003287 optical effect Effects 0.000 title claims abstract description 296
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000013145 classification model Methods 0.000 claims abstract description 64
- 238000012549 training Methods 0.000 claims abstract description 22
- 239000013598 vector Substances 0.000 claims description 42
- 239000000835 fiber Substances 0.000 claims description 25
- 238000012360 testing method Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 12
- 238000010998 test method Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 8
- 238000002474 experimental method Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000002411 adverse Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000013307 optical fiber Substances 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000004020 conductor Substances 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 208000002173 dizziness Diseases 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 229920002430 Fibre-reinforced plastic Polymers 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000011151 fibre-reinforced plastic Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000010189 synthetic method Methods 0.000 description 1
Landscapes
- Optical Communication System (AREA)
Abstract
The invention provides a fault optical cable prediction method, a device, electronic equipment and a storage medium, which are used for realizing the prediction of whether an optical cable to be predicted is a fault optical cable or not through an optical cable classification model and improving the prediction efficiency. The target sample set adopted during the optical cable classification model training is obtained by adjusting the sample unbalance rate of the initial sample set by utilizing the quantity of distinguishing parameter characteristics among optical cable samples carrying category labels in the initial sample set to obtain the target unbalance rate and adjusting the first quantity of the optical cable samples carrying the target category labels in the initial sample set by utilizing the target unbalance rate, so that the classification accuracy of the optical cable classification model obtained through training can be improved, and the prediction accuracy of a fault optical cable is further improved.
Description
Technical Field
The present invention relates to the field of optical fiber communications technologies, and in particular, to a method and apparatus for predicting a faulty optical cable, an electronic device, and a storage medium.
Background
With the continuous development of information technology, the optical fiber communication technology is increasingly widely applied in the information age of everything interconnection. The optical cable network serves as a basic physical network of the Internet and plays an irreplaceable supporting role for various service networks. The reliable operation of the optical cable network is ensured, and the smooth transmission of various information is ensured. For this reason, it is important to perform a faulty cable prediction, i.e., to determine whether the cable is a faulty cable.
In the prior art, when the fault optical cable is predicted, an optical cable classification model is generally adopted for implementation. The cable classification model needs to be trained with a sample set containing parameter characteristics corresponding to faulty cable samples and non-faulty cable samples prior to application.
However, as the optical cable faults are small probability events, the number of faulty optical cable samples in the sample set is obviously smaller than that of non-faulty optical cable samples, the number of samples in different categories in the sample set is extremely unbalanced, the classification accuracy of the optical cable classification model obtained through training is reduced, and the accurate prediction of faulty optical cables is not facilitated.
Disclosure of Invention
The invention provides a fault optical cable prediction method, a device, electronic equipment and a storage medium, which are used for solving the defects in the prior art.
The invention provides a fault optical cable prediction method, which comprises the following steps:
acquiring parameter characteristic values of the optical cable to be predicted;
inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted is a fault optical cable or not, wherein the prediction result is output by the optical cable classification model;
the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample;
The category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
According to the fault optical cable prediction method provided by the invention, the method further comprises the following steps:
constructing feature vectors corresponding to the parameter features based on the parameter feature values corresponding to the optical cable samples, and determining label vectors of the optical cable samples;
determining the correlation between the feature vector and the label vector by adopting a Pearson test method;
and selecting the distinguishing parameter characteristic from the parameter characteristics based on the correlation.
According to the method for predicting the faulty fiber optic cable provided by the invention, the distinguishing parameter features are selected from the parameter features based on the correlation, and the method comprises the following steps:
for any parameter feature, calculating test statistics of t distribution based on correlation between a feature vector corresponding to the any parameter feature and the tag vector;
and if the test statistic is smaller than a preset threshold value, determining that any parameter characteristic is the distinguishing parameter characteristic.
According to the method for predicting the faulty optical cable provided by the invention, the sample unbalance rate of the initial sample set is adjusted based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in the initial sample set, so as to obtain a target unbalance rate, and the method comprises the following steps:
determining a penalty term based on the number of distinguishing parameter features;
and adjusting the sample unbalance rate based on the punishment item to obtain the target unbalance rate.
According to the method for predicting the faulty fiber cable provided by the invention, the sample unbalance rate is adjusted based on the penalty term to obtain the target unbalance rate, and the method comprises the following steps:
and calculating the difference value between the sample unbalance rate and the punishment item to obtain the target unbalance rate.
According to the method for predicting the faulty optical cable provided by the invention, the adjusting the first number of optical cable samples carrying the target class labels in the initial sample set based on the target unbalance rate includes:
calculating the equivalent number of the optical cable samples carrying the non-fault labels in the initial sample set based on the target unbalance rate;
and determining a second number of optical cable samples carrying the fault labels after adjustment based on the equivalent number, and adding the second number of optical cable samples carrying the fault labels in the initial sample set.
According to the method for predicting the fault optical cable provided by the invention, the method for obtaining the parameter characteristic value of the optical cable to be predicted comprises the following steps:
acquiring non-numerical parameter characteristic information of the optical cable to be predicted;
and performing One-Hot coding on the non-numerical parameter characteristics to obtain parameter characteristic values corresponding to the non-numerical parameter characteristic information.
The invention also provides a fault optical cable prediction device, which comprises:
the acquisition module is used for acquiring the parameter characteristic value of the optical cable to be predicted;
the prediction module is used for inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted output by the optical cable classification model is a fault optical cable;
the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample;
The category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the fault optical cable prediction method according to any one of the above when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a faulty fiber optic cable prediction method as described in any of the above.
The invention also provides a computer program product comprising a computer program which when executed by a processor implements a faulty cable prediction method according to any one of the above.
According to the method, the device, the electronic equipment and the storage medium for predicting the fault optical cable, the prediction of whether the optical cable to be predicted is the fault optical cable or not is realized through the optical cable classification model, and the prediction efficiency can be improved. The target sample set adopted during the optical cable classification model training is obtained by adjusting the sample unbalance rate of the initial sample set by utilizing the quantity of distinguishing parameter characteristics among optical cable samples carrying category labels in the initial sample set to obtain the target unbalance rate and adjusting the first quantity of the optical cable samples carrying the target category labels in the initial sample set by utilizing the target unbalance rate, so that the classification accuracy of the optical cable classification model obtained through training can be improved, and the prediction accuracy of a fault optical cable is further improved. Moreover, the target unbalance rate is introduced to adjust the first quantity, so that compared with the scheme of directly adjusting the first quantity through the quantity difference of the optical cable samples of different types of labels in the initial sample set, the quantity of the optical cable samples of the increased target type labels is smaller, the time required for balancing the optical cable samples can be reduced, and the adverse effect of the increased optical cable samples on the performance of the optical cable classification model can be reduced.
Drawings
In order to more clearly illustrate the invention or the technical solutions of 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 to those skilled in the art that other drawings can be obtained according to these drawings without inventive effort.
FIG. 1 is a flow chart of a method for predicting a faulty fiber optic cable provided by the present invention;
FIG. 2 is a schematic diagram of the variation of the unbalance rate IR with the number of samples p of the optical cable according to the present invention;
FIG. 3 is a schematic diagram of a faulty fiber optic cable prediction apparatus provided by the present invention;
fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Because the optical cable faults are small probability events, the number of faulty optical cable samples in the sample set for training the optical cable classification model is obviously smaller than that of non-faulty optical cable samples, which leads to the extremely unbalanced number of samples in different categories in the sample set, so that the classification accuracy of the optical cable classification model obtained by training is reduced, and the accurate prediction of the faulty optical cable is not facilitated.
Currently, the conventional index for measuring the class imbalance between a faulty cable sample and a non-faulty cable sample is the imbalance rate IR, i.e. the ratio of the number of non-faulty cable samples (majority class) to the number of faulty cable samples (minority class). IR is defined as:
wherein N is maj For the number of non-faulty cable samples, N min Is the number of failed cable samples. It is apparent that when ir=1, the sample set is a perfectly balanced set. When IR>The greater the IR, the greater the degree of imbalance of the sample set at 1.
In sample sets with an unbalanced sample number, two sample sets with the same unbalance rate but different cable sample numbers may have very different classification performance. That is, although the two sample sets have the same unbalance rate, since the sample sets having more distinguishing features have better classification performance, they cannot be said to have the same unbalance degree. Also, in performing faulty cable predictions, when the ratio of faults to non-faults of two types of cables is consistent, and IR is used to measure, the same sample balancing method is used for both types of cables, i.e. the corresponding number of samples is increased or decreased. In practice, however, because one type of fiber optic cable has more distinguishing characteristics between faults and non-faults than the other type of fiber optic cable, it can take less time to perform a sample balancing process than the sample set of the other type of fiber optic cable. Therefore, the embodiment of the invention provides a fault optical cable prediction method, which adjusts the unbalance rate in the sample set, so as to adjust the number of fault optical cable samples in the sample set, and further make the number of optical cable samples carrying different types of labels in the sample set more balanced.
Fig. 1 is a flowchart of a fault optical cable prediction method provided in an embodiment of the present invention, as shown in fig. 1, where the method includes:
s1, acquiring parameter characteristic values of an optical cable to be predicted;
s2, inputting the parameter characteristic values of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted output by the optical cable classification model is a fault optical cable;
the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample;
the category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
Specifically, the execution main body of the fault optical cable prediction method provided in the embodiment of the present invention is a fault optical cable prediction device, and the device may be configured in a server, where the server may be a local server, or may be a cloud server, and the local server may be a computer, or the like.
Firstly, step S1 is executed, and parameter characteristic values of the optical cable to be predicted are obtained. The optical cable to be predicted refers to an optical cable for which whether the optical cable is a fault optical cable needs to be determined. The parameter characteristic value refers to a numerical representation of a parameter characteristic, and the parameter characteristic can comprise hundreds of rated tensile strength, bending radius, thickness of an outer sheath of the optical cable, dielectric strength of the outer sheath, insulation resistance of the outer sheath, armoured and metal reinforced cores of the dielectric strength in the outer sheath, line level, manufacturer, outer diameter, radius of a traction end, traction length, traction tension, reserved length, dizziness date, laying type, category, number of fiber cores in the optical cable, structural mode, transmission conductor, medium condition and the like.
The parameter features may include a numerical parameter feature and a non-numerical parameter feature, for which a value may be directly obtained, and for a score execution parameter feature, information may be intelligently obtained, and at this time, the information may be converted into a form of a value by a conventional method, and the method adopted in the conversion is not particularly limited.
And then executing step S2, inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model, analyzing the parameter characteristic value of the optical cable to be predicted by adopting the optical cable classification model, and obtaining and outputting a prediction result of whether the optical cable to be predicted is a fault optical cable. The prediction result may include that the optical cable to be predicted is a fault optical cable and that the optical cable to be predicted is not a fault optical cable, where the two prediction results may be marked separately, for example, the optical cable to be predicted is a fault optical cable and may be marked as 1, the optical cable to be predicted is not a fault optical cable and may be marked as 0, or may be marked in other manners.
It is understood that the optical cable classification model used may be a machine learning model, for example, a support vector machine model, a decision tree model, or a k-nearest neighbor algorithm model. The cable classification model may be trained from a target sample set that may be determined from an initial sample set that is a sample set prior to adjustment by the target imbalance rate and a target sample set that is a sample set after adjustment by the target imbalance rate.
In the initial sample set, parameter feature values corresponding to the cable samples carrying the category labels may be included, the category labels may include fault labels and non-fault labels, and the cable samples may include fault cable samples carrying the fault labels and non-fault cable samples carrying the non-fault labels. The sample imbalance ratio of the initial sample set can be calculated by the above-mentioned IR calculation formula.
The distinguishing parameter features are parameter features for distinguishing optical cable samples carrying different types of labels, play a key role in distinguishing different types of optical cable samples in an unbalanced sample set, and the sample set with more distinguishing parameter features has better classification performance. The distinguishing parameter characteristics may include at least 15 of a bend radius, a line level, a manufacturer, an outer diameter, a pulling end radius, a pulling length, a pulling tension, a reserved length, a dizziness date, a lay type, a category, a number of fiber cores in the optical cable, a structural mode, a transmission conductor, a medium condition, and the like of the optical cable. The rated tensile strength, the thickness of the outer sheath, the dielectric strength of the outer sheath, the insulation resistance of the outer sheath, the armor of the dielectric strength in the outer sheath, the metal reinforcing core and the like of the optical cable in the parameter characteristics are all indistinguishable parameter characteristics.
For example, there are two sample sets A, B, the sample capacity distributions of the two different classes are the same. The same degree of class imbalance will then be obtained using existing metrics, since these only consider the distribution of classes. Now, if a adds more distinguishing parameter features than B, it is expected that the classification performance of a will be better than B, as the added distinguishing parameter features introduce valuable distinguishing information.
When different distinguishing parameter characteristics are added in the unbalanced sample set, the change condition and the change trend of the classification performance can be proved through the following experiments.
Simulation experiment one: experiments were performed at a variety of different cable sample numbers, different IR. Wherein, the value range of the optical cable sample number p is {2, 10, 50, 100, 500, 1000}, and the value range of the IR is {5, 10, 50, 100, 500}. That is, given IR, sample set classification performance was studied for 6 different p values. In the experiment, normal distribution N (. Mu.) was used respectively maj ,∑ maj ) And N (mu) min ,∑ min ) Representing both majority class and minority class distributions. It is considered that μ is a constant variance condition min Sum mu maj The larger the difference between them, the more discriminating the features. Four comparative experiments can be performed as follows: 1) P-2 non-distinguishing parameter features are added for both p-dimensional minority data and majority data; 2) Adding (p-2) 10% non-distinguishing parameter features for both p-dimensional minority data and majority data; 3) Adding (p-2) 50% non-distinguishing parameter features for both p-dimensional minority data and majority data; 4) Both p-dimensional minority class data and majority class data are added (p-2) 90% of the non-distinguishing parameter features.
If the number of the non-distinguishing parameter features to be added is an integer, and if the calculated value is a decimal, it is necessary to round up.
Simulation experiment II: the distinguishing parameter features and the non-distinguishing parameter features of different mixing ratios are added. In real world data, features are typically not discriminative or indistinguishable as in simulation experiments one. Here, data in which distinguishing parameter characteristics and non-distinguishing parameter characteristics are mixed are simulated, and variations in the distinguishing performance are studied in which the mixing ratio is varied. The experimental setup was still to perform experiments at a variety of different cable sample numbers and different IR. Wherein, the value range of the optical cable sample number p is {2, 10,50, 100, 500, 1000}, and the value range of the IR is {5, 10,50, 100, 500}. That is, given IR, sample set classification performance was studied for 6 different p values. K% of distinguishing parameter features and (100-k)% of non-distinguishing parameter features can be added in the experiment, wherein the value range of k is as follows: {10,50,90}.
The following conclusions can be drawn from the above experiments:
1. for sample sets with the same IR but different cable sample numbers p, the classification performance is different, and considering the negative correlation between the unbalance rate and the classification performance, it is not appropriate to use the same IR to describe the unbalance rate of these sample sets.
2. The greater the number of distinguishing parameter features, the better the classification performance. Because the distinguishing parameter features can bring more distinguishing information, the difference between the categories is increased, and better classification is performed.
Therefore, the sample unbalance rate of the initial sample set can be adjusted through the number of distinguishing parameter characteristics among the optical cable samples carrying the category labels in the initial sample set, and the target unbalance rate is obtained.
The distinguishing parameter features may be determined by correlation between the parameter features corresponding to the optical cable samples and the class labels, or may be directly given, which is not specifically limited herein.
Thereafter, the first number of cable samples carrying the target class labels in the initial sample set may be adjusted to obtain a target sample set by the target imbalance. The target class labels are fault labels, namely, the first number of fault optical cable samples in the initial sample set is required to be adjusted through the target unbalance rate, the obtained optical cable samples carrying different class labels in the target sample set are equivalent in number, the optical cable samples can be identical or different, and the optical cable samples carrying different class labels have identical contribution to the training of the optical cable classification model.
The method for predicting the fault optical cable provided by the embodiment of the invention comprises the steps of firstly obtaining parameter characteristic values of the optical cable to be predicted; and then inputting the parameter characteristic value of the optical cable to be predicted into the optical cable classification model to obtain a prediction result of whether the optical cable to be predicted output by the optical cable classification model is a fault optical cable. The prediction of whether the optical cable to be predicted is a fault optical cable or not is realized through the optical cable classification model, so that the prediction efficiency can be improved. The target sample set adopted during the optical cable classification model training is obtained by adjusting the sample unbalance rate of the initial sample set by utilizing the quantity of distinguishing parameter characteristics among optical cable samples carrying category labels in the initial sample set to obtain the target unbalance rate and adjusting the first quantity of the optical cable samples carrying the target category labels in the initial sample set by utilizing the target unbalance rate, so that the classification accuracy of the optical cable classification model obtained through training can be improved, and the prediction accuracy of a fault optical cable is further improved. Moreover, the target unbalance rate is introduced to adjust the first quantity, so that compared with the scheme of directly adjusting the first quantity through the quantity difference of the optical cable samples of different types of labels in the initial sample set, the quantity of the optical cable samples of the increased target type labels is smaller, the time required for balancing the optical cable samples can be reduced, and the adverse effect of the increased optical cable samples on the performance of the optical cable classification model can be reduced.
The adverse effect is that the method for adding the optical cable sample is usually to synthesize or resample the parameter characteristic value corresponding to the optical cable sample carrying the target class label, and the optical cable sample obtained in this way can amplify the characteristic of the optical cable sample carrying the target class label in the target sample set, thereby affecting the performance of the optical cable classification model.
Based on the foregoing embodiment, the method for predicting a faulty fiber optic cable provided in the embodiment of the present invention further includes:
constructing feature vectors corresponding to the parameter features based on the parameter feature values corresponding to the optical cable samples, and determining label vectors of the optical cable samples;
determining the correlation between the feature vector and the label vector by adopting a Pearson test method;
and selecting the distinguishing parameter characteristic from the parameter characteristics based on the correlation.
Specifically, in the embodiment of the invention, the distinguishing parameter characteristic can be determined by a Pearson test. Using Pearson's test, the correlation between two feature vectors can be effectively detected. If the correlation between a feature vector and a label vector is non-zero, then this parameter feature can be considered as a distinguishing parameter feature.
When distinguishing parameter characteristics are determined, characteristic vectors corresponding to the parameter characteristics can be constructed according to parameter characteristic values corresponding to the optical cable samples, and label vectors corresponding to the characteristic vectors can be determined.
Assuming that there are N cable samples in the initial sample set, each cable sample has p parameter features, the initial sample set can be expressed asWherein x is i ＝[x 1i ,x 2i ,...,x pi ] T ∈R p×1 And y is i E { -1,1}. Defining the feature vector corresponding to the j-th parameter feature as x j ＝[x j1 ,x j2 ,...,x jN ] T (j=1, 2,.,. P.), the label vector of the cable sample is y= [ y 1 ,y 2 ,...,y N ]。
Then, determining the characteristic direction by using a Pearson test methodQuantity x j And tag vector y= [ y ] 1 ,y 2 ,...,y N ]The correlation between the two can be expressed by using a Pearson correlation coefficient, and the Pearson correlation coefficient can be expressed as ρ j 。
Thereafter, by correlation, a distinguishing parameter feature may be selected from among the parameter features, and the number of distinguishing parameter features may be determined. In the embodiment of the invention, all the parameter characteristics corresponding to the characteristic vectors with correlation with the label vector can be used as distinguishing parameter characteristics.
In the embodiment of the invention, the distinguishing parameter characteristics are determined by the Pearson test method, so that the accuracy of the distinguishing parameter characteristics can be improved, the performance of the optical cable classification model is further improved, and the accuracy of the prediction result is improved.
On the basis of the foregoing embodiment, in the method for predicting a faulty optical cable according to the embodiment of the present invention, the selecting, based on the correlation, the distinguishing parameter feature from the parameter features includes:
for any parameter feature, calculating test statistics of t distribution based on correlation between a feature vector corresponding to the any parameter feature and the tag vector;
and if the test statistic is smaller than a preset threshold value, determining that any parameter characteristic is the distinguishing parameter characteristic.
Specifically, in the embodiment of the present invention, when the distinguishing parameter feature is selected from the parameter features through correlation, for any parameter feature, the test statistic of t distribution may be calculated through correlation between the feature vector corresponding to any parameter feature and the tag vector.
Taking any parameter feature as the j-th parameter feature as an example, the feature vector corresponding to any parameter feature is the j-th feature vector x j Test statistic t of t distribution j Can be expressed as:
wherein t is j Obeying a t-distribution with degrees of freedom N-2.
Thereafter, the magnitude relation between the test statistic and a preset threshold value is judged, and the preset threshold value can be set according to the requirement, for example, the preset threshold value can be set as |t N-2,1-α | a. The invention relates to a method for producing a fibre-reinforced plastic composite. If the test statistic is smaller than the preset threshold, t is present j ＜|t N-2,1-α I, reject the original hypothesis H at this time 0 Select alternative hypothesis H 1 I.e. consider the feature vector x j The corresponding parameter features are distinguishing parameter features. And we mean the number of distinguishing features in a class. Either parameter characteristic may be determined to be a distinguishing parameter characteristic.
For each of the parameter features, the above procedure is performed, i.e. Pearson test is applied to each parameter feature, and H is counted 0 Number of times p of refusal * All the distinguishing parameter features can be selected from the parameter features.
Here, H 0 Number of times p of refusal * I.e. the number of distinguishing parameter features.
In the embodiment of the invention, the determination of the distinguishing parameter characteristics is realized by calculating the test statistic of the t distribution, so that the determination process can be simplified, and the accuracy of the number of the obtained distinguishing parameter characteristics can be improved.
Based on the foregoing embodiment, the method for predicting a faulty optical cable according to the embodiment of the present invention adjusts a sample imbalance rate of an initial sample set based on the number of distinguishing parameter features between optical cable samples carrying category labels in the initial sample set, to obtain a target imbalance rate, including:
Determining a penalty term based on the number of distinguishing parameter features;
and adjusting the sample unbalance rate based on the punishment item to obtain the target unbalance rate.
Specifically, in the embodiment of the invention, in the process of obtaining the target unbalance rate, the punishment item can be determined by the number of distinguishing parameter features. The penalty term can be expressed as: λlog (p) * )。Wherein is the parameter controlling the importance of the penalty term.
The penalty term may adjust the impact of the number of cable samples in the initial sample set on the classification performance of the trained cable classification model. In particular, in certain extreme cases, there may be p * =0, in which case log (p * ) And are not defined. To solve this problem, at p * Will p when=0 * Set to p * ＝1。
And then, the sample unbalance rate of the initial sample set can be adjusted according to the penalty term, namely, the penalty term is introduced on the basis of the sample unbalance rate of the initial sample set, and the target unbalance rate IR' is obtained.
It is apparent that for an initial sample set with a fixed IR, the adjusted IR' decreases with increasing distinguishing parameter characteristics. Thus, the tuned IR' has a better negative correlation with the classification performance of the trained cable classification model, since the classification performance of the cable classification model follows p * And become better with an increase in (c).
In the embodiment of the invention, the punishment items are constructed through the quantity of distinguishing parameter characteristics, so that the target unbalance rate is obtained, and the target unbalance rate is lower than the initial sample unbalance rate, therefore, the first quantity is regulated by combining the target unbalance rate, the unbalance degree of the initial sample set can be counteracted under the condition that the regulation amplitude of the first quantity is smaller, and the unbalance degree of the target sample set is 0.
On the basis of the foregoing embodiment, in the method for predicting a faulty optical cable according to the embodiment of the present invention, the adjusting the sample imbalance rate based on the penalty term to obtain the target imbalance rate includes:
and calculating the difference value between the sample unbalance rate and the punishment item to obtain the target unbalance rate.
Specifically, the target unbalance rate may be expressed as:
IR′＝IR-λlog(p * )
in the embodiment of the invention, the target unbalance rate is determined by calculating the difference value between the sample unbalance rate and the punishment item, so that the calculation process of the target unbalance rate can be simplified, and the efficiency is improved.
On the basis of the foregoing embodiment, the method for predicting a faulty optical cable according to the embodiment of the present invention adjusts, based on the target imbalance rate, a first number of optical cable samples carrying a target class label in the initial sample set, including:
Calculating the equivalent number of the optical cable samples carrying the non-fault labels in the initial sample set based on the target unbalance rate;
and determining a second number of optical cable samples carrying the fault labels after adjustment based on the equivalent number, and adding the second number of optical cable samples carrying the fault labels in the initial sample set.
Specifically, in the embodiment of the present invention, when the first number of optical cable samples carrying the target class labels in the initial sample set is adjusted, the equivalent number of optical cable samples carrying the non-fault labels in the initial sample set may be calculated according to the target unbalance rate.
Let x be the equivalent number of cable samples carrying non-faulty labels in the initial sample set, and N be the first number of cable samples carrying target class labels in the initial sample set min The following steps are:
then there are:
x＝IR′*N min
thereafter, the equivalent number x may be directly taken as the second number of cable samples adjusted to carry the faulty label, and the second number of cable samples carrying the faulty label may be added to the initial sample set. The adding means may be a SMOTE method, for example, a synthetic method or a resampling method, which is not particularly limited herein.
For example, if the initial sample set involves 3852 cable samples, of which there are 201 faulty cable samples and 3651 non-faulty cable samples, the faulty samples only account for 5.2% of the total samples, and the sample imbalance ratio is:
the target unbalance rate obtained after adjustment is as follows:
the equivalent number is:
x＝IR′*N min ＝16.99*201＝3415
and then, the parameter characteristic values corresponding to 3415 fault optical cable samples can be supplemented in the initial sample set, and a target sample set can be obtained. The number of failed cable samples and the number of non-failed cable samples each account for about 50% of the target sample set, with the data being balanced.
If the number difference between the fault optical cable samples and the non-fault optical cable samples in the initial sample set is directly based, the parameter characteristic values corresponding to 3450 fault optical cable samples are required to be added into the initial sample set. It can be seen that the method provided in the embodiment of the present invention can reduce the number of optical cable samples added to the initial sample set.
It can be understood that in the target sample set, 7267 parameter feature values corresponding to the optical cable samples are combined, and 80% of parameter feature values corresponding to the optical cable samples can be randomly selected from the parameter feature values to form a training sample, and the remaining 20% of parameter feature values corresponding to the optical cable samples form a test sample.
In the embodiment of the invention, a specific method for determining the second number is provided, so that the number of samples required to be increased can be reduced, and the adverse effect of the increased samples on the performance of the optical classification model is reduced.
On the basis of the foregoing embodiments, the method for predicting a faulty optical cable according to the embodiment of the present invention includes:
acquiring non-numerical parameter characteristic information of the optical cable to be predicted;
and performing One-Hot coding on the non-numerical parameter characteristics to obtain parameter characteristic values corresponding to the non-numerical parameter characteristic information.
Specifically, in the embodiment of the invention, when the parameter characteristic value of the optical cable to be predicted is obtained, the non-numerical parameter characteristic information of the optical cable to be predicted can be obtained first, wherein the non-numerical parameter characteristic information refers to a characteristic which cannot be directly represented by a numerical value, and at the moment, the non-numerical parameter characteristic can be subjected to One-Hot coding to obtain the parameter characteristic value corresponding to the non-numerical parameter characteristic information. Namely, through One-Hot coding, the non-numerical parameter characteristic information can be converted into numerical parameter characteristic values.
One-Hot encoding is a representation of the classification variables as binary vectors. This first requires mapping the classification value to an integer value. Each integer value is then represented as a binary vector, which is zero except for the index of the integer, which is labeled 1.
As shown in fig. 2, the unbalance rate IR varies with the number p of optical cable samples. In fig. 2, curve 1 represents ir=100, curve 2 represents the sample imbalance rate, and curve 3 represents the target imbalance rate. As the number of cable samples in the sample set increases, the IR' after adjustment gradually decreases. This means that the degree of imbalance of the sample set is reduced, thereby reducing the number of sample samples during the over-sampling or under-sampling process.
After the target sample set is obtained, the optical cable classification model can be obtained by training the training samples in the target sample set, and the prediction results of the optical cable classification model obtained by using the test samples are shown in the following table:
TABLE 1 prediction results for cable classification models
| Number of faulty fiber optic cable samples | Number of non-faulty fiber optic cable samples | |
| True value | 41 | 676 |
| Predictive value | 33 | 684 |
In the embodiment of the invention, the prediction result of the optical fiber classification model is measured by adopting the F1 score (F1-score), wherein the F1-score is a measurement index of the classification problem, is a harmonic mean of the precision rate and the recall rate, and has the maximum value of 1 and the minimum value of 0, and the expression is as follows:
where precision represents the accuracy rate (representing how many of the cable samples predicted to be non-faulty cable are true non-faulty cable samples), and recovery represents the recall rate (representing how many of the cable samples were predicted to be faulty by the cable classification model).
For the sample unbalance rate IR before unbalance rate adjustment of 18.16, the target unbalance rate IR' after adjustment of 16.99, and the result value of the F1-score after adjustment of 71.89%, which is improved by 0.86% compared with the result value of the F1-score before adjustment, the pretreatment method for predicting the faulty optical cable, which is established in the embodiment of the invention, not only reduces the time of oversampling samples in the unbalanced sample treatment process, but also improves the prediction accuracy of the faulty optical cable under the condition of not increasing the software and hardware cost. Therefore, the optical cable classification model in the embodiment of the invention can provide a certain auxiliary effect for the prediction of the fault optical cable.
The information age has arrived, and various industries are developing in relation to communication technology nowadays. Therefore, the communication level requirements of enterprises and people are high, and the development speed of the communication industry is very rapid. The quality of the communication technology is directly related to the maintenance of the current communication optical cable, and the optical cable classification model in the embodiment of the invention not only can be used for detecting faults of the communication optical cable, but also can be used for predicting various signal faults, such as signal problems in the Beidou positioning process and signal faults in the 5G transmission process, can be effectively predicted through the optical cable classification model in the embodiment of the invention, so that personnel can be reminded to process the faults, safety accidents are reduced, and the invention has certain commercial value in the aspects of signal fault prediction, optical cable fault prediction and the like.
As shown in fig. 3, on the basis of the above embodiment, in an embodiment of the present invention, there is provided a fault optical cable prediction apparatus, including:
the obtaining module 31 is configured to obtain a parameter characteristic value of the optical cable to be predicted;
the prediction module 32 is configured to input the parameter feature value of the optical cable to be predicted to an optical cable classification model, so as to obtain a prediction result of whether the optical cable to be predicted output by the optical cable classification model is a faulty optical cable;
the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample;
the category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
On the basis of the above embodiment, the fault optical cable prediction device provided in the embodiment of the present invention further includes a distinguishing parameter feature selection module, configured to:
constructing feature vectors corresponding to the parameter features based on the parameter feature values corresponding to the optical cable samples, and determining label vectors of the optical cable samples;
determining the correlation between the feature vector and the label vector by adopting a Pearson test method;
and selecting the distinguishing parameter characteristic from the parameter characteristics based on the correlation.
On the basis of the foregoing embodiments, the device for predicting a faulty optical cable provided in the embodiment of the present invention is specifically configured to:
for any parameter feature, calculating test statistics of t distribution based on correlation between a feature vector corresponding to the any parameter feature and the tag vector;
and if the test statistic is smaller than a preset threshold value, determining that any parameter characteristic is the distinguishing parameter characteristic.
On the basis of the foregoing embodiment, the device for predicting a faulty optical cable provided in the embodiment of the present invention further includes an adjustment module, configured to:
determining a penalty term based on the number of distinguishing parameter features;
And adjusting the sample unbalance rate based on the punishment item to obtain the target unbalance rate.
On the basis of the foregoing embodiments, the device for predicting a faulty optical cable provided in the embodiment of the present invention is specifically configured to:
and calculating the difference value between the sample unbalance rate and the punishment item to obtain the target unbalance rate.
On the basis of the foregoing embodiments, the device for predicting a faulty optical cable provided in the embodiment of the present invention is further configured to:
calculating the equivalent number of the optical cable samples carrying the non-fault labels in the initial sample set based on the target unbalance rate;
and determining a second number of optical cable samples carrying the fault labels after adjustment based on the equivalent number, and adding the second number of optical cable samples carrying the fault labels in the initial sample set.
On the basis of the foregoing embodiments, the device for predicting a faulty optical cable provided in the embodiment of the present invention is specifically configured to:
acquiring non-numerical parameter characteristic information of the optical cable to be predicted;
and performing One-Hot coding on the non-numerical parameter characteristics to obtain parameter characteristic values corresponding to the non-numerical parameter characteristic information.
Specifically, the functions of each module in the fault optical cable prediction device provided in the embodiment of the present invention are in one-to-one correspondence with the operation flows of each step in the above method embodiment, and the achieved effects are consistent.
Fig. 4 illustrates a physical schematic diagram of an electronic device, as shown in fig. 4, which may include: processor (Processor) 410, communication interface (Communications Interface) 420, memory (Memory) 430, and communication bus 440, wherein Processor 410, communication interface 420, and Memory 430 communicate with each other via communication bus 440. Processor 410 may invoke logic instructions in memory 430 to perform the faulty fiber optic cable prediction method provided in the embodiments described above, the method comprising: acquiring parameter characteristic values of the optical cable to be predicted; inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted is a fault optical cable or not, wherein the prediction result is output by the optical cable classification model; the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample; the category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program, when executed by a processor, being capable of executing the fault cable prediction method provided in the above embodiments, the method comprising: acquiring parameter characteristic values of the optical cable to be predicted; inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted is a fault optical cable or not, wherein the prediction result is output by the optical cable classification model; the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample; the category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the faulty fiber optic cable prediction method provided in the above embodiments, the method including: acquiring parameter characteristic values of the optical cable to be predicted; inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted is a fault optical cable or not, wherein the prediction result is output by the optical cable classification model; the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample; the category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the 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 scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A method of predicting a faulty fiber optic cable, comprising:
acquiring parameter characteristic values of the optical cable to be predicted;
inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted is a fault optical cable or not, wherein the prediction result is output by the optical cable classification model;
the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample;
The category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
2. The method of predicting a faulty fiber optic cable of claim 1, wherein the method further comprises:
constructing feature vectors corresponding to the parameter features based on the parameter feature values corresponding to the optical cable samples, and determining label vectors of the optical cable samples;
determining the correlation between the feature vector and the label vector by adopting a Pearson test method;
and selecting the distinguishing parameter characteristic from the parameter characteristics based on the correlation.
3. The method of claim 2, wherein selecting the distinguishing parameter feature from the parameter features based on the correlation comprises:
for any parameter feature, calculating test statistics of t distribution based on correlation between a feature vector corresponding to the any parameter feature and the tag vector;
And if the test statistic is smaller than a preset threshold value, determining that any parameter characteristic is the distinguishing parameter characteristic.
4. The method for predicting a faulty fiber optic cable according to claim 1, wherein the adjusting the sample imbalance rate of the initial sample set based on the number of distinguishing parameter features between fiber optic cable samples carrying class labels in the initial sample set to obtain the target imbalance rate includes:
determining a penalty term based on the number of distinguishing parameter features;
and adjusting the sample unbalance rate based on the punishment item to obtain the target unbalance rate.
5. The method for predicting a faulty fiber optic cable of claim 4, wherein said adjusting the sample imbalance rate based on the penalty term to obtain the target imbalance rate comprises:
and calculating the difference value between the sample unbalance rate and the punishment item to obtain the target unbalance rate.
6. The method of claim 1-5, wherein adjusting the first number of cable samples in the initial sample set that carry a target class label based on the target imbalance rate comprises:
Calculating the equivalent number of the optical cable samples carrying the non-fault labels in the initial sample set based on the target unbalance rate;
and determining a second number of optical cable samples carrying the fault labels after adjustment based on the equivalent number, and adding the second number of optical cable samples carrying the fault labels in the initial sample set.
7. The method for predicting a faulty fiber optic cable of any one of claims 1-5, wherein the obtaining the parameter characteristic value of the fiber optic cable to be predicted includes:
acquiring non-numerical parameter characteristic information of the optical cable to be predicted;
and performing One-Hot coding on the non-numerical parameter characteristics to obtain parameter characteristic values corresponding to the non-numerical parameter characteristic information.
8. A faulty fiber optic cable prediction apparatus comprising:
the acquisition module is used for acquiring the parameter characteristic value of the optical cable to be predicted;
the prediction module is used for inputting the parameter characteristic value of the optical cable to be predicted into an optical cable classification model to obtain a prediction result of whether the optical cable to be predicted output by the optical cable classification model is a fault optical cable;
the optical cable classification model is obtained based on training of a target sample set; the target sample set is obtained through the following steps: based on the number of distinguishing parameter characteristics among optical cable samples carrying category labels in an initial sample set, adjusting the sample unbalance rate of the initial sample set to obtain a target unbalance rate, and based on the target unbalance rate, adjusting the first number of the optical cable samples carrying the target category labels in the initial sample set to obtain the optical cable sample;
The category labels comprise fault labels or non-fault labels, the target category labels are fault labels, the initial sample set contains parameter characteristic values corresponding to the optical cable samples, and the distinguishing parameter characteristics are parameter characteristics for distinguishing the optical cable samples carrying different category labels.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the faulty fiber optic cable prediction method of any one of claims 1 to 7 when the program is executed by the processor.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the faulty fiber optic cable prediction method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210398229.XA CN116975674A (en) | 2022-04-15 | 2022-04-15 | Fault optical cable prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210398229.XA CN116975674A (en) | 2022-04-15 | 2022-04-15 | Fault optical cable prediction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116975674A true CN116975674A (en) | 2023-10-31 |
Family
ID=88480075
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210398229.XA Pending CN116975674A (en) | 2022-04-15 | 2022-04-15 | Fault optical cable prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116975674A (en) |
-
2022
- 2022-04-15 CN CN202210398229.XA patent/CN116975674A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104331635B (en) | The method of power optical fiber Communication ray power prediction | |
| CN111583012B (en) | Method for evaluating default risk of credit, debt and debt main body by fusing text information | |
| CN112651849B (en) | Method and system for identifying abnormal voltage monitoring data based on unbalanced data set | |
| CN115145812B (en) | Test case generation method and device, electronic equipment and storage medium | |
| CN107291775A (en) | The reparation language material generation method and device of error sample | |
| CN111738017A (en) | Intention identification method, device, equipment and storage medium | |
| CN114418409A (en) | Equipment safety risk assessment method and device based on multiple neural networks | |
| CN116975674A (en) | Fault optical cable prediction method and device, electronic equipment and storage medium | |
| CN116522912B (en) | Training method, device, medium and equipment for package design language model | |
| US7797136B2 (en) | Metrics to evaluate process objects | |
| CN115616470B (en) | Method, system, equipment and medium for predicting metering error state of current transformer | |
| CN112231299A (en) | Method and device for dynamically adjusting feature library | |
| CN113468816B (en) | Training method and device for iron tower steel index prediction model and readable storage medium | |
| CN115953022A (en) | Elevator safety evaluation method and system, electronic device and storage medium | |
| CN113258682B (en) | Intelligent control method and system for optimizing operation of transformer | |
| CN115129019A (en) | Training method of production line fault analysis model and production line fault analysis method | |
| CN115526393A (en) | Construction cost prediction method based on transformer project key influence factor screening | |
| CN117407796B (en) | Cross-component small sample fault diagnosis method, system and storage medium | |
| CN115879783A (en) | Key performance index abnormity detection method and device, electronic equipment and storage medium | |
| CN113850292A (en) | Failure knowledge base generation type updating method, system, electronic equipment and medium | |
| CN113705973A (en) | Neural network learning-based power grid security risk online evaluation method | |
| CN115134294B (en) | Method and device for determining standby route and computer readable storage medium | |
| CN115759828A (en) | Power transmission line state evaluation method based on intention recognition slot filling joint algorithm | |
| CN116503125A (en) | Article recommendation method, device, equipment and medium | |
| CN116740526A (en) | Unbalanced ship coating defect image generation method, system and readable 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 |