CN112039687A - Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network - Google Patents

Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network Download PDF

Info

Publication number
CN112039687A
CN112039687A CN202010673514.9A CN202010673514A CN112039687A CN 112039687 A CN112039687 A CN 112039687A CN 202010673514 A CN202010673514 A CN 202010673514A CN 112039687 A CN112039687 A CN 112039687A
Authority
CN
China
Prior art keywords
network
data
discriminator
generator
training
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
Application number
CN202010673514.9A
Other languages
Chinese (zh)
Inventor
朱晓荣
曹家明
池德盛
陈雨萱
沈雍钧
田忆军
庄益康
史坤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
JIANGSU EAST CENTURY NETWORK INFORMATION Co.,Ltd.
MOX AUTOMATIC CONTROL INFORMATION (CHINA) Co.,Ltd.
Original Assignee
Jiangsu East Century Network Information Co ltd
Nanjing University of Posts and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiangsu East Century Network Information Co ltd, Nanjing University of Posts and Telecommunications filed Critical Jiangsu East Century Network Information Co ltd
Priority to CN202010673514.9A priority Critical patent/CN112039687A/en
Publication of CN112039687A publication Critical patent/CN112039687A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/16—Threshold monitoring

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Environmental & Geological Engineering (AREA)
  • Monitoring And Testing Of Transmission In General (AREA)

Abstract

The invention discloses a small sample characteristic-oriented generation-based countermeasure network fault diagnosis method, which solves the problems that in the fault detection and diagnosis process, the cost of manually adding labels to network data is too high, the convergence fluctuation of a generated countermeasure network is large, the label loss in actually collected network data is serious, and the like; firstly, analyzing the cause of network fault, and providing a semi-supervised fault diagnosis model by improving the loss function of a generator network and the output layer of a discriminator network; then, the model is further optimized, an algorithm for combining a generation countermeasure network and a convolutional neural network is provided, wherein the generation countermeasure network is responsible for generating data of various fault categories, and then the data are used for training the convolutional neural network to finish the diagnosis of network faults; the method can realize accurate diagnosis of network faults under the condition of a small amount of labeled data.

Description

Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network
Technical Field
The invention relates to the technical field of communication networks, in particular to a fault diagnosis method for a small sample characteristic based on an improved generation countermeasure network.
Background
With the coming of big data era and the rapid development of technologies such as deep learning, people can utilize a complex neural network model to mine and extract key information in mass data under the support of strong calculation power. Especially in a complex heterogeneous network environment, thousands of network nodes generate a large amount of network operation information every day, and under the development trend of network convergence and isomerization, fault diagnosis is a key research direction. Fault diagnosis is one of the main tasks to manage any network.
The traditional network fault diagnosis is mainly to compare alarm information of network performance indexes with an expert experience base and manually analyze and investigate faults, but in the heterogeneous wireless network environment with large scale and complex structure, a diagnosis mode based on manpower analysis can occupy a large amount of manpower and material resources and increase maintenance cost, so that a dynamic and self-adaptive network fault diagnosis method is urgently needed, accurate detection and diagnosis of network faults in the complex network environment can be realized, the hazards of service interruption, network paralysis and the like caused by fault propagation are effectively relieved, great significance is brought to the evolution of a wireless network, and the exploration and research of a more efficient and more intelligent fault diagnosis technology in the heterogeneous network is bound to become one of important subjects of future heterogeneous network research.
In recent years, the generation of confrontation networks as a typical method for realizing artificial intelligence has been widely successful in the fields of computer vision, image recognition and natural language processing, and people have gained miraculous ability in the aspect of processing complex problems. It is a new framework for estimating the generation network through the countermeasure process, where two models are trained simultaneously: a generative model G to capture the data distribution, and a discriminant model D to estimate the likelihood of samples from the training data rather than the model G. The training scheme for model G is to maximize the probability that model D makes a mistake, this framework is relevant to the very least two-party game. The potential of the framework is proved by the experimental result through identifying the minist handwriting data set by utilizing the framework. At present, the idea of generating an anti-network is mainly applied to the fields of computer vision and image recognition, and what can be done by the "magic" technology for wireless communication systems? This is a considerable problem.
The invention provides a method for applying a countermeasure network generating idea to the field of network fault detection and diagnosis and combining the countermeasure network generating idea with a typical network fault diagnosis method. By utilizing the thought of generating the confrontation network, a large number of reliable data sets with marks are obtained for training the network fault diagnosis algorithm, and the problems that historical data obtained from a real network is not rich enough, and the effect of constructing a diagnosis system is not ideal are solved. By the method, time for manually marking the training data is greatly saved, and the precision of the fault diagnosis model is improved.
Disclosure of Invention
The purpose of the invention is as follows: the invention provides a fault diagnosis method for small sample characteristics based on an improved generation countermeasure network, which can save the time for manually marking training data and improve the precision of a fault diagnosis model.
The technical scheme is as follows: in order to achieve the purpose, the invention adopts the technical scheme that:
a fault diagnosis method for a small sample feature based on an improved generation countermeasure network comprises the following steps:
step S1, collecting a network state data set with a label from a heterogeneous wireless network environment, and selecting partial network parameters to identify a fault cell through an algorithm combining Relieff and mutual information; the specific selection method is as follows:
s1.1, calculating the weight of each feature, namely the network performance index, by using a Relieff, and sequencing according to the weight to obtain a set A1;
s1.2, calculating mutual information of the features in the set A1 pairwise, and deleting the features with smaller weight when the mutual information is larger than a preset threshold value to obtain a set A2;
s1.3, selecting the features larger than the artificially set weight threshold value from the set A2 as neural network parameters of a detection stage;
step S2, respectively carrying out normalization processing on a small number of collected network state data sets with labels based on the maximum value of each key performance indicator KPI;
s3, establishing and generating a confrontation network model according to the neural network parameters defined in the step S1, and dividing a data set with a mark into a training set and a test set;
step S4, inputting the training set in the step S3 into a generated confrontation network model according to fault categories, wherein a generator and a discriminator structure of the generated confrontation network model select a convolutional neural network; after the model training is converged, generating data of corresponding fault categories through a generator network, adding fault labels, and merging the network state data set with the labels, the training set and the generated data in the step S1 to serve as new training data of the convolutional neural network diagnosis model;
and step S5, inputting the new training data acquired in the step S4 into the improved SGAN fault detection model for training to obtain a diagnosis result.
Further, the normalization processing manner in step S2 is as follows:
Figure BDA0002583202500000021
wherein
Figure BDA0002583202500000022
The ith key performance index after normalization; max (KPI)i) The maximum value of the ith key performance index; normalized network state
Figure BDA0002583202500000023
The method specifically comprises the following steps:
Figure BDA0002583202500000031
further, the generating the confrontation network model is based on an improved semi-supervised generation confrontation network model, wherein the loss function of the final generator G is:
Figure BDA0002583202500000032
wherein λ is a specific gravity coefficient; x is the real data of the image data,
Figure BDA0002583202500000033
is to generate data, xrIs a sample, x, in a small real datasetfGenerating samples in the dataset for the minibatch, n being the size of the minibatch dataset, l being the number of hidden layers in the arbiter, f () representing the output of the hidden layers of the arbiter, E () representing the expected value of the distribution function,
Figure BDA0002583202500000034
presentation generation data
Figure BDA0002583202500000035
Probability of being discriminated as class K +1 by the discriminator D.
Further, the specific steps of generating the confrontation network training data based on the improved semi-supervised in step S4 are as follows:
s4.1, adopting two fully-connected neural networks to respectively form a generator and a discriminator;
s4.2, training the generator to enable the generator to respectively simulate a small number of different network state data sets with labels collected in a heterogeneous wireless network environment to generate simulation data;
s4.3, respectively inputting a small amount of network state data sets with labels collected from the heterogeneous wireless network environment and simulation data generated by the generator into a discriminator to train a discriminator D; keeping the parameters of the generator G unchanged, updating the parameters of the discriminator D as follows:
Figure BDA0002583202500000036
where n is the size of the small batch dataset,
Figure BDA0002583202500000037
representing generated data
Figure BDA0002583202500000038
Probability of being discriminated as class K +1 by the discriminator D;
step S4.4, training the generator G, keeping the parameters of the discriminator D unchanged, and updating the parameters of the generator G as shown in the following:
Figure BDA0002583202500000041
wherein z is(i)Representing the noise samples when training the generator, g (z) representing the generating function;
s4.5, performing alternate iterative training on the generator G and the discriminator D until the model converges to realize balance;
s4.6, generating a simulation data set with labels and representing different network states;
and S4.7, acquiring the trained discriminator D.
Further, the step S5 of generating the anti-network fault detection model based on improved semi-supervision specifically includes:
the network output layer of the discriminator adopts a softmax classifier to realize classification of real data; when the discriminator structure is changed, the loss function of the generator G is as follows:
Figure BDA0002583202500000042
wherein
Figure BDA0002583202500000043
It is the generator that generates the data that,
Figure BDA0002583202500000044
is that
Figure BDA0002583202500000045
Is determined by the probability distribution function of (a),
Figure BDA0002583202500000046
presentation generation data
Figure BDA0002583202500000047
Probability of being discriminated as class K +1 by the discriminator D;
the loss function of the discriminator D is divided into supervised and unsupervised losses, as follows:
Figure BDA0002583202500000048
where x is the truth, y is the class label for x, p (x, y) is the probability distribution for (x, y) obedience,
Figure BDA0002583202500000049
it is the generator that generates the data that,
Figure BDA00025832025000000413
is the probability distribution to which it is subjected; l issupRepresents a supervised loss section, LunsupIs an unsupervised loss fraction;
adding feature matching and compactness calculation in a loss function of a generator network, and using the feature matching and compactness calculation to stabilize convergence of the model and improve the final performance of the model; objective function for increasing distance between generated data and real data in loss function of generator G
Figure BDA00025832025000000410
The formula is as follows:
Figure BDA00025832025000000411
wherein f isi() Is the output of the i-th hidden layer of the discriminator, l is the number of hidden layers, x is the real data,
Figure BDA00025832025000000412
is to generate data;
increasing a distance function of the compactness between the real samples and the compactness between the generated samples in a loss function of the generator; the distance between samples is a standardized Euclidean distance, and the distance function of the real sample compactness and the generated sample compactness is shown as follows:
Figure BDA0002583202500000051
wherein Dreal，DfakeRespectively representing the real inter-sample compactness and the generated inter-sample compactness,
Figure BDA0002583202500000052
is input data of a discriminator, xr,xfRespectively a small-batch real data set and a small-batch generated data set, n is the size of the small-batch data set, and f () is the middle hiding of the discriminatorOutput of layers, | |)normalIs the normalized euclidean distance;
the loss function of the final generator G is shown as follows:
Figure BDA0002583202500000053
further, the trigger condition for setting model correction for the trained discriminator model is as follows:
when the discriminator carries out network fault diagnosis, the data which are wrongly diagnosed are stored in the database; triggering retraining of the model when the stored data is greater than one-third of the training data; wherein the training data consists of the original training data and the data for diagnosing the error.
Has the advantages that:
the invention provides a small sample characteristic-oriented generation-based confrontation network fault diagnosis method, which solves the problem that the effect of constructing a diagnosis system is not ideal due to insufficient historical data obtained from a real network. By the method, not only is the time for manually marking the training data saved greatly, but also the precision of the fault diagnosis model is improved. .
Drawings
Fig. 1 is a heterogeneous wireless network scenario diagram;
FIG. 2 is a flow chart of the improved generation-based countermeasure network fault diagnosis for a small sample feature provided by the present invention;
fig. 3 is a diagram of an improved semi-supervised generated countermeasure network model provided by the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings.
In the invention, a dense heterogeneous wireless network scene with a multilayer network structure is formed by a high-power macro base station and a low-power micro base station shown in fig. 1, and in the scene, due to the diversity of the network, the system becomes more complex and the network management becomes more difficult. The invention considers the network fault detection and diagnosis in the scene, and firstly screens out useful network parameters aiming at the reasons that the specific network scene analysis may cause the fault, which is the work that needs to be done in the early stage of building the network fault diagnosis model. And then acquiring historical data from a heterogeneous wireless network historical database, wherein the historical data comprises a fault category variable set, a fault variable set and key performance indicators KPI thereof.
Based on the heterogeneous wireless network scenario shown in fig. 1, the invention provides an improved generation countermeasure network fault diagnosis method facing to small sample characteristics, which comprises the following specific steps:
step S1, collecting a network state data set with a label from a heterogeneous wireless network environment, and selecting partial network parameters to identify a fault cell through an algorithm combining Relieff and mutual information; the specific selection method is as follows:
s1.1, calculating the weight of each feature, namely the network performance index, by using a Relieff, and sequencing according to the weight to obtain a set A1;
s1.2, calculating mutual information of the features in the set A1 pairwise, and deleting the features with smaller weight when the mutual information is larger than a preset threshold value to obtain a set A2;
s1.3, selecting the features larger than a manually preset weight threshold value from the set A2 as neural network parameters of a detection stage;
step S2, respectively carrying out normalization processing on a small number of collected network state data sets with labels based on the maximum value of each key performance indicator KPI; the specific normalization processing mode is as follows:
Figure BDA0002583202500000061
wherein
Figure BDA0002583202500000062
The ith key performance index after normalization; max (KPI)i) The maximum value of the ith key performance index; normalized network state
Figure BDA0002583202500000063
The method specifically comprises the following steps:
Figure BDA0002583202500000064
s3, establishing and generating a confrontation network model according to the neural network parameters defined in the step S1, and dividing a data set with a mark into a training set and a test set; the generation of the countermeasure network model is based on an improved semi-supervised generation countermeasure network model, wherein the loss function of the final generator G is as follows:
Figure BDA0002583202500000065
wherein, λ is a specific gravity coefficient used for adjusting the specific gravity between the original SGAN cross entropy loss function and the new objective function; x is the real data of the image data,
Figure BDA0002583202500000071
is to generate data, xrIs a sample, x, in a small real datasetfGenerating samples in the dataset for the minibatch, n being the size of the minibatch dataset, l being the number of hidden layers in the arbiter, f () representing the output of the hidden layers of the arbiter, E () representing the expected value of the distribution function,
Figure BDA0002583202500000072
representing generated data
Figure BDA0002583202500000073
Probability of being discriminated as class K +1 by the discriminator D.
Step S4, inputting the training set in the step S3 into a generated confrontation network model according to fault categories, wherein a generator and a discriminator structure of the generated confrontation network model select a convolutional neural network; after the model training is converged, generating data of corresponding fault categories through a generator network, adding fault labels, and merging the network state data set with the labels, the training set and the generated data in the step S1 to serve as new training data of the convolutional neural network diagnosis model;
the specific steps of generating the confrontation network training data based on the improved semi-supervision are as follows:
s4.1, adopting two fully-connected neural networks to respectively form a generator and a discriminator;
s4.2, training the generator to enable the generator to respectively simulate a small number of different network state data sets with labels collected in a heterogeneous wireless network environment to generate simulation data;
s4.3, respectively inputting a small amount of network state data sets with labels collected from the heterogeneous wireless network environment and simulation data generated by the generator into a discriminator to train a discriminator D; keeping the parameters of the generator G unchanged, updating the parameters of the discriminator D as follows:
Figure BDA0002583202500000074
where n is the size of the small batch dataset,
Figure BDA0002583202500000075
representing generated data
Figure BDA0002583202500000076
Probability of being discriminated as class K +1 by the discriminator D;
step S4.4, training the generator G, keeping the parameters of the discriminator D unchanged, and updating the parameters of the generator G as shown in the following:
Figure BDA0002583202500000077
wherein z is(i)Representing the noise samples when training the generator, g (z) representing the generating function;
s4.5, performing alternate iterative training on the generator G and the discriminator D until the model converges to realize balance;
s4.6, generating a simulation data set with labels and representing different network states;
and S4.7, acquiring the trained discriminator D.
And step S5, inputting the new training data acquired in the step S4 into the improved SGAN fault detection model for training to obtain a diagnosis result.
Specifically, a softmax classifier is adopted by a network output layer of the discriminator to realize classification of real data; when the structure of the discriminator is changed, the loss function of the generator G is as follows:
Figure BDA0002583202500000081
wherein
Figure BDA0002583202500000082
It is the generator that generates the data that,
Figure BDA0002583202500000083
is that
Figure BDA0002583202500000084
Is determined by the probability distribution function of (a),
Figure BDA0002583202500000085
presentation generation data
Figure BDA0002583202500000086
Probability of being discriminated as class K +1 by the discriminator D;
the loss function of the discriminator D is divided into supervised and unsupervised losses, as follows:
Figure BDA0002583202500000087
where x is the truth, y is the class label for x, p (x, y) is the probability distribution for (x, y) obedience,
Figure BDA0002583202500000088
it is the generator that generates the data that,
Figure BDA0002583202500000089
is the probability distribution to which it is subjected; l issupRepresents a supervised loss section, LunsupIs an unsupervised loss fraction;
adding feature matching and compactness calculation in a loss function of a generator network, and using the feature matching and compactness calculation to stabilize convergence of the model and improve the final performance of the model; objective function for increasing distance between generated data and real data in loss function of generator G
Figure BDA00025832025000000810
The formula is as follows:
Figure BDA00025832025000000811
wherein f isi() Is the output of the i-th hidden layer of the discriminator, l is the number of hidden layers, x is the real data,
Figure BDA00025832025000000812
is to generate data;
increasing a distance function of the compactness between the real samples and the compactness between the generated samples in a loss function of the generator; the distance between samples is a standardized Euclidean distance, and the distance function of the real sample compactness and the generated sample compactness is shown as follows:
Figure BDA00025832025000000813
wherein Dreal，DfakeRespectively representing the real inter-sample compactness and the generated inter-sample compactness,
Figure BDA00025832025000000814
is input data of a discriminator, xr,xfAre respectivelyA small batch of real data sets and samples in the small batch of generated data sets, n being the size of the small batch of data sets, f () being the output of the discriminator intermediate hidden layer, | |)normalIs the normalized euclidean distance;
the loss function of the final generator G is shown as follows:
Figure BDA0002583202500000091
when the discriminator carries out network fault diagnosis, the data which are wrongly diagnosed are stored in the database; triggering retraining of the model when the stored data is greater than one-third of the training data; wherein the training data consists of the original training data and the data for diagnosing the error.
To illustrate the effectiveness of the proposed method of the present invention, an example is given below. The collection of example data is generated by a dynamic heterogeneous wireless network environment implemented in OPNET. A cellular network consisting of 3 macro base stations and 15 micro base stations is built by using OPNET simulation software, the whole coverage area of the network is 5km multiplied by 5km, the coverage radius of each macro base station is 1km, 5 micro base stations are distributed in the range of each macro base station, users are randomly distributed in respective cells, and specific parameters are shown in the following table 1:
TABLE 1 simulation parameter Table for OPNET heterogeneous wireless network
Figure BDA0002583202500000092
In the network simulation, 5 different types of faults are mainly set in the chapter, namely uplink interference, downlink interference, coverage fault, base station fault and link fault, and no fault state. Additional consideration was given to 16 key performance indicators, as shown in table 2 below. Before the simulation begins, the time for occurrence and recovery of the faults is preset so as to manually add fault labels to data generated by the simulation, the simulation is set for 24 hours, the total duration of each fault is 2 hours, and the fault is recovered after 30 minutes of each time. Example a method for improved generation of fault diagnosis of a countermeasure network using data collected by OPNET, comprising the steps of:
step 1: an input vector form is determined. If a network fault FC occurs in a certain period of time TiThen, the network status during this time period is specifically:
Figure BDA0002583202500000101
step 2: and for convenient analysis, the KPI data is subjected to normalization processing. The normalization processing mode specifically comprises the following steps:
Figure BDA0002583202500000102
Figure BDA0002583202500000103
the i-th key performance index after normalization is indicated. max (KPI)i) Refers to the maximum value at which the ith key performance indicator occurs.
Normalized network state
Figure BDA0002583202500000104
The method specifically comprises the following steps: step 1
Figure BDA0002583202500000105
And step 3: establishing a generation confrontation network model according to defined neural network parameters, dividing marked real data into training data and test data according to a certain proportion, selecting the generation confrontation network model as an improved semi-supervised generation confrontation network model, and improving a loss function of a final generator G of the SGAN as follows:
Figure BDA0002583202500000106
where λ is a weight coefficient used to adjust the weight between the original SGAN cross entropy loss function and the new objective function, x is the real data,
Figure BDA0002583202500000107
is to generate data, xr,xfThe method comprises the steps of respectively generating samples in a small-batch real data set and a small-batch generated data set, wherein n is the size of the small-batch data set, and l is the number of hidden layers in a discriminator.
And 4, step 4: and (3) inputting training data into the generated countermeasure network model one by one according to fault category classification, generating data of corresponding fault categories through a generator network after the model training is converged, adding fault labels, and combining the training data and the generated data to be used as new training data of the convolutional neural network diagnosis model. The key performance indicators are shown in table 2 below:
TABLE 2 OPNET heterogeneous wireless network key performance index
Figure BDA0002583202500000111
And finally, eight key performance indexes including RSRQ _ P, DCR, HO, RSRP _ P, ERAB _ S, SNR _ UL, SNR _ DL and LER are selected for fault diagnosis.
And 5: and dividing the preprocessed data set into a training set and a testing set, inputting the training set and the testing set into an improved SGAN fault detection model for training, and obtaining a diagnosis result.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.

Claims (6)

1. A fault diagnosis method for a small sample characteristic based on an improved generation countermeasure network is characterized by comprising the following steps:
step S1, collecting a network state data set with a label from a heterogeneous wireless network environment, and selecting partial network parameters to identify a fault cell through an algorithm combining Relieff and mutual information; the specific selection method is as follows:
s1.1, calculating the weight of each feature, namely the network performance index, by using a Relieff, and sorting according to the weight to obtain a set A1;
s1.2, calculating mutual information of the features in the set A1 pairwise, and deleting the features with smaller weight when the mutual information is larger than a preset threshold value to obtain a set A2;
s1.3, selecting the features larger than a manually preset weight threshold value from the set A2 as neural network parameters of a detection stage;
step S2, respectively carrying out normalization processing on a small number of collected network state data sets with labels based on the maximum value of each key performance indicator KPI;
s3, establishing and generating a confrontation network model according to the neural network parameters defined in the step S1, and dividing a data set with marks into a training set and a test set;
step S4, inputting the training set in the step S3 into a generated confrontation network model according to fault categories, wherein a generator and a discriminator structure of the generated confrontation network model select a convolutional neural network; after the model training is converged, generating data of corresponding fault categories through a generator network, adding fault labels, and merging the network state data set with the labels, the training set and the generated data in the step S1 to serve as new training data of the convolutional neural network diagnosis model;
and step S5, inputting the new training data obtained in the step S4 into an improved SGAN fault detection model for training to obtain a diagnosis result.
2. The method for diagnosing the fault of the small sample feature based on the improved generation countermeasure network according to claim 1, wherein the normalization processing in the step S2 is as follows:
Figure FDA0002583202490000011
wherein
Figure FDA0002583202490000012
The ith key performance index after normalization; max (KPI)i) The maximum value of the ith key performance index; normalized network state
Figure FDA0002583202490000013
The method specifically comprises the following steps:
Figure FDA0002583202490000014
3. the fault diagnosis method for the small sample feature based on the improved generation countermeasure network as claimed in claim 1, wherein the generation countermeasure network model is an improved semi-supervised generation countermeasure network model, and the loss function of the final generator G is:
Figure FDA0002583202490000021
wherein λ is a specific gravity coefficient; x is the real data of the image data,
Figure FDA0002583202490000022
is to generate data, xrIs a sample, x, in a small real datasetfGenerating samples in the dataset for the minibatch, n being the size of the minibatch dataset, l being the number of hidden layers in the arbiter, f () representing the output of the hidden layers of the arbiter, E () representing the expected value of the distribution function,
Figure FDA0002583202490000027
representing generated data
Figure FDA0002583202490000028
Probability of being discriminated as class K +1 by the discriminator D.
4. The method for fault diagnosis of small sample feature based on improved generation of countermeasure network as claimed in claim 3, wherein the specific steps of generating countermeasure network training data based on improved semi-supervised in step S4 are as follows:
s4.1, adopting two fully-connected neural networks to respectively form a generator and a discriminator;
s4.2, training the generator to enable the generator to respectively simulate a small number of different network state data sets with labels collected in a heterogeneous wireless network environment to generate simulation data;
s4.3, respectively inputting a small amount of network state data sets with labels collected from the heterogeneous wireless network environment and simulation data generated by the generator into a discriminator to train a discriminator D; keeping the parameters of the generator G unchanged, updating the parameters of the discriminator D as follows:
Figure FDA0002583202490000023
where n is the size of the small batch dataset,
Figure FDA0002583202490000024
representing generated data
Figure FDA0002583202490000025
Probability of being discriminated as class K +1 by the discriminator D;
step S4.4, training the generator G, keeping the parameters of the discriminator D unchanged, and updating the parameters of the generator G as shown in the following:
Figure FDA0002583202490000026
wherein z is(i)To representNoise samples when training the generator, g (z) representing the generating function;
s4.5, performing alternate iterative training on the generator G and the discriminator D until the model converges to realize balance;
s4.6, generating a simulation data set with labels and representing different network states;
and S4.7, acquiring the trained discriminator D.
5. The method for diagnosing faults of a small sample-oriented antagonistic network based on improved generation is claimed in claim 1, wherein the fault detection model of the antagonistic network based on improved semi-supervised generation in step S5 is specifically:
the network output layer of the discriminator adopts a softmax classifier to realize classification of real data; when the structure of the discriminator is changed, the loss function of the generator G is as follows:
Figure FDA0002583202490000031
wherein
Figure FDA0002583202490000032
It is the generator that generates the data that,
Figure FDA0002583202490000033
is that
Figure FDA0002583202490000034
Is determined by the probability distribution function of (a),
Figure FDA0002583202490000035
representing generated data
Figure FDA0002583202490000036
Probability of being discriminated as class K +1 by the discriminator D;
the loss function of the discriminator D is divided into supervised and unsupervised losses, as follows:
Figure FDA0002583202490000037
where x is the truth, y is the class label for x, p (x, y) is the probability distribution for (x, y) obedience,
Figure FDA0002583202490000038
it is the generator that generates the data that,
Figure FDA0002583202490000039
is the probability distribution to which it is subjected; l issupRepresents a supervised loss section, LunsupIs an unsupervised loss part;
adding feature matching and compactness calculation in a loss function of a generator network, and using the feature matching and the compactness calculation to stabilize convergence of the model and improve the final performance of the model; objective function for increasing distance between generated data and real data in loss function of generator G
Figure FDA00025832024900000310
The formula is as follows:
Figure FDA00025832024900000311
wherein f isi() Is the output of the i-th hidden layer of the discriminator, l is the number of hidden layers, x is the real data,
Figure FDA00025832024900000312
is to generate data;
increasing a distance function of the compactness between the real samples and the compactness between the generated samples in a loss function of the generator; the distance between the samples adopts a standardized Euclidean distance, and the distance function of the real sample compactness and the generated sample compactness is as follows:
Figure FDA0002583202490000041
wherein Dreal，DfakeRespectively representing the real inter-sample compactness and the generated inter-sample compactness,
Figure FDA0002583202490000042
is input data of a discriminator, xr,xfIs the small batch of real data sets and the small batch of samples in the generated data sets, respectively, n is the size of the small batch of data sets, f () is the output of the middle hidden layer of the discriminator, | |normalIs the normalized euclidean distance;
the loss function of the final generator G is shown as follows:
Figure FDA0002583202490000043
6. the fault diagnosis method for the small sample feature based on the improved generation countermeasure network as claimed in claim 5, wherein the trigger condition for setting the model modification for the trained discriminator model is as follows:
when the discriminator carries out network fault diagnosis, the data which are wrongly diagnosed are stored in the database; triggering retraining of the model when the stored data is greater than one-third of the training data; wherein the training data consists of the original training data and the data for diagnosing errors.
CN202010673514.9A 2020-07-14 2020-07-14 Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network Pending CN112039687A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010673514.9A CN112039687A (en) 2020-07-14 2020-07-14 Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010673514.9A CN112039687A (en) 2020-07-14 2020-07-14 Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network

Publications (1)

Publication Number Publication Date
CN112039687A true CN112039687A (en) 2020-12-04

Family

ID=73579551

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010673514.9A Pending CN112039687A (en) 2020-07-14 2020-07-14 Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network

Country Status (1)

Country Link
CN (1) CN112039687A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112597702A (en) * 2020-12-21 2021-04-02 电子科技大学 Pneumatic modeling generation type confrontation network model training method based on radial basis function
CN112702760A (en) * 2020-12-16 2021-04-23 西安电子科技大学 Method, system, medium, equipment, terminal and application for estimating cell load
CN113159127A (en) * 2021-03-23 2021-07-23 浙江大胜达包装股份有限公司 Intelligent fault diagnosis method for gravure printing machine guide roller based on improved generation countermeasure network
CN113485863A (en) * 2021-07-14 2021-10-08 北京航空航天大学 Method for generating heterogeneous unbalanced fault samples based on improved generation countermeasure network
CN113721113A (en) * 2021-09-02 2021-11-30 广西大学 Fault line selection method based on semi-supervised generation countermeasure network
CN114124676A (en) * 2021-11-19 2022-03-01 南京邮电大学 Fault root cause positioning method and system for network intelligent operation and maintenance system
CN114501525A (en) * 2022-01-28 2022-05-13 东南大学 Wireless network interruption detection method based on condition generation countermeasure network
CN114895647A (en) * 2022-04-13 2022-08-12 哈尔滨工程大学 Small-sample ship part fault data-oriented diagnosis method and readable storage medium
CN114970628A (en) * 2022-05-26 2022-08-30 青岛酒店管理职业技术学院 Rotating part fault diagnosis method based on generation countermeasure network under condition of unbalanced samples
CN115127192A (en) * 2022-05-20 2022-09-30 中南大学 Semi-supervised water chilling unit fault diagnosis method and system based on graph neural network

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110617966A (en) * 2019-09-23 2019-12-27 江南大学 Bearing fault diagnosis method based on semi-supervised generation countermeasure network
CN110823574A (en) * 2019-09-30 2020-02-21 安徽富煌科技股份有限公司 Fault diagnosis method based on semi-supervised learning deep countermeasure network
CN111242225A (en) * 2020-01-16 2020-06-05 南京邮电大学 Fault detection and diagnosis method based on convolutional neural network

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110617966A (en) * 2019-09-23 2019-12-27 江南大学 Bearing fault diagnosis method based on semi-supervised generation countermeasure network
CN110823574A (en) * 2019-09-30 2020-02-21 安徽富煌科技股份有限公司 Fault diagnosis method based on semi-supervised learning deep countermeasure network
CN111242225A (en) * 2020-01-16 2020-06-05 南京邮电大学 Fault detection and diagnosis method based on convolutional neural network

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112702760B (en) * 2020-12-16 2022-03-15 西安电子科技大学 Method, system, medium, equipment, terminal and application for estimating cell load
CN112702760A (en) * 2020-12-16 2021-04-23 西安电子科技大学 Method, system, medium, equipment, terminal and application for estimating cell load
CN112597702A (en) * 2020-12-21 2021-04-02 电子科技大学 Pneumatic modeling generation type confrontation network model training method based on radial basis function
CN112597702B (en) * 2020-12-21 2022-07-19 电子科技大学 Pneumatic modeling generation type confrontation network model training method based on radial basis function
CN113159127A (en) * 2021-03-23 2021-07-23 浙江大胜达包装股份有限公司 Intelligent fault diagnosis method for gravure printing machine guide roller based on improved generation countermeasure network
CN113485863B (en) * 2021-07-14 2023-05-05 北京航空航天大学 Method for generating heterogeneous imbalance fault samples based on improved generation of countermeasure network
CN113485863A (en) * 2021-07-14 2021-10-08 北京航空航天大学 Method for generating heterogeneous unbalanced fault samples based on improved generation countermeasure network
CN113721113A (en) * 2021-09-02 2021-11-30 广西大学 Fault line selection method based on semi-supervised generation countermeasure network
CN113721113B (en) * 2021-09-02 2023-10-24 广西大学 Fault line selection method based on semi-supervised generation countermeasure network
CN114124676A (en) * 2021-11-19 2022-03-01 南京邮电大学 Fault root cause positioning method and system for network intelligent operation and maintenance system
CN114124676B (en) * 2021-11-19 2024-04-02 南京邮电大学 Fault root positioning method and system for network intelligent operation and maintenance system
CN114501525A (en) * 2022-01-28 2022-05-13 东南大学 Wireless network interruption detection method based on condition generation countermeasure network
CN114501525B (en) * 2022-01-28 2024-02-02 东南大学 Wireless network interruption detection method based on condition generation countermeasure network
CN114895647A (en) * 2022-04-13 2022-08-12 哈尔滨工程大学 Small-sample ship part fault data-oriented diagnosis method and readable storage medium
CN115127192A (en) * 2022-05-20 2022-09-30 中南大学 Semi-supervised water chilling unit fault diagnosis method and system based on graph neural network
CN115127192B (en) * 2022-05-20 2024-01-23 中南大学 Semi-supervised water chilling unit fault diagnosis method and system based on graph neural network
CN114970628A (en) * 2022-05-26 2022-08-30 青岛酒店管理职业技术学院 Rotating part fault diagnosis method based on generation countermeasure network under condition of unbalanced samples

Similar Documents

Publication Publication Date Title
CN112039687A (en) Small sample feature-oriented fault diagnosis method based on improved generation countermeasure network
CN110929918B (en) 10kV feeder fault prediction method based on CNN and LightGBM
CN109142171B (en) Urban PM10 concentration prediction method based on feature expansion and fusing with neural network
CN109492026B (en) Telecommunication fraud classification detection method based on improved active learning technology
CN112685504B (en) Production process-oriented distributed migration chart learning method
CN108537259A (en) Train control on board equipment failure modes and recognition methods based on Rough Sets Neural Networks model
CN111586728B (en) Small sample characteristic-oriented heterogeneous wireless network fault detection and diagnosis method
CN114580706A (en) Power financial service wind control method and system based on GRU-LSTM neural network
CN104777418A (en) Analog circuit fault diagnosis method based on depth Boltzman machine
CN113709779A (en) Cellular network fault diagnosis method
CN113225346A (en) Network operation and maintenance situation assessment method based on machine learning
CN113553624A (en) WGAN-GP privacy protection system and method based on improved PATE
Wu et al. Unsupervised deep transfer learning for fault diagnosis in fog radio access networks
CN115345222A (en) Fault classification method based on TimeGAN model
CN105160598A (en) Power grid service classification method based on improved EM algorithm
CN114897085A (en) Clustering method based on closed subgraph link prediction and computer equipment
CN114553671A (en) Diagnosis method for power communication network fault alarm
CN110335160A (en) A kind of medical treatment migratory behaviour prediction technique and system for improving Bi-GRU based on grouping and attention
CN115659258B (en) Power distribution network fault detection method based on multi-scale graph roll-up twin network
CN116431988A (en) Resident trip activity time sequence generation method based on activity mode-Markov chain
CN116031879A (en) Hybrid intelligent feature selection method suitable for transient voltage stability evaluation of power system
CN115734274A (en) Cellular network fault diagnosis method based on deep learning and knowledge graph
CN114265954B (en) Graph representation learning method based on position and structure information
CN115293249A (en) Power system typical scene probability prediction method based on dynamic time sequence prediction
CN114861792A (en) Complex power grid key node identification method based on deep reinforcement learning

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20210701

Address after: No.130 Qunxian Road, Jiaxing, Zhejiang 314000

Applicant after: MOX AUTOMATIC CONTROL INFORMATION (CHINA) Co.,Ltd.

Applicant after: JIANGSU EAST CENTURY NETWORK INFORMATION Co.,Ltd.

Address before: 210046 No. 9 Wenyuan Road, Qixia District, Nanjing, Jiangsu Province

Applicant before: NANJING University OF POSTS AND TELECOMMUNICATIONS

Applicant before: JIANGSU EAST CENTURY NETWORK INFORMATION Co.,Ltd.

AD01 Patent right deemed abandoned
AD01 Patent right deemed abandoned

Effective date of abandoning: 20230707