CN112862211A - Method and device for assigning orders of dynamic ring defects of communication management system - Google Patents

Method and device for assigning orders of dynamic ring defects of communication management system Download PDF

Info

Publication number
CN112862211A
CN112862211A CN202110254521.XA CN202110254521A CN112862211A CN 112862211 A CN112862211 A CN 112862211A CN 202110254521 A CN202110254521 A CN 202110254521A CN 112862211 A CN112862211 A CN 112862211A
Authority
CN
China
Prior art keywords
defect
management system
communication management
neural network
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
CN202110254521.XA
Other languages
Chinese (zh)
Inventor
万莹
许鸿飞
张翼
吕海军
张少军
邢宁哲
张继东
刘昀
刘天丽
王彦璋
来媛
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.)
State Grid Corp of China SGCC
Information and Telecommunication Branch of State Grid Jibei Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
Information and Telecommunication Branch of State Grid Jibei Electric Power Co Ltd
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 State Grid Corp of China SGCC, Information and Telecommunication Branch of State Grid Jibei Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202110254521.XA priority Critical patent/CN112862211A/en
Publication of CN112862211A publication Critical patent/CN112862211A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Primary Health Care (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • Medical Informatics (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention discloses a communication management system dynamic ring defect order assigning method and a device, wherein the method comprises the following steps: collecting the dynamic ring alarm data of the communication management system; inputting the collected dynamic ring alarm data into a trained defect fault prediction model, and outputting the defect fault information of a communication management system, wherein the fault prediction model is obtained by performing machine learning training on a deep residual error neural network; and generating the maintenance work order information to be distributed according to the defect fault information of the communication management system. The method is based on the deep residual error neural network training defect fault prediction model, can predict the defect fault by utilizing large-scale dynamic ring alarm data, greatly improves the accuracy rate of fault defect prediction, further generates a corresponding maintenance work order according to the predicted defect fault information, and can greatly improve the practicability, intelligence and high efficiency of automatic dispatching work of the dynamic ring defect work order.

Description

Method and device for assigning orders of dynamic ring defects of communication management system
Technical Field
The invention relates to the field of communication management systems, in particular to a method and a device for assigning orders of dynamic ring defects of a communication management system.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
The power communication network is one of indispensable important components of a power system, is the basis of power grid dispatching automation and production management modernization, and is an important technical means for ensuring safe, economic and stable operation of a power grid. In the information communication operation and maintenance site, in the face of a huge and complex information set, how to conveniently and quickly acquire the fault information of the power communication network is the only solution for quickly completing the maintenance task and improving the maintenance efficiency.
With the popularization of machine learning algorithms in various industries, the prior art realizes the prediction of fault information of the power communication network by means of the machine learning algorithms, but because the power communication network is a complex network, the faults of the power communication network are various (for example, laser performance degradation, overhigh or overlow working environment temperature, overhigh or overlow optical power, single board fault, tail fiber performance degradation, splice box fault, optical fiber core interruption, optical fiber interruption and the like), although the data dimension is not high, the attributes are complex, and influence and association relations exist among the single board fault, the data dimension and the tail fiber performance are not uniform in different environments, and difficulty is brought to the selection of early-stage features of machine learning. Moreover, because the rules for feature extraction are often formulated based on artificial priori knowledge, and mining analysis is lacked, some unaware associations are difficult to find. In addition, when equipment faults in the power communication network are predicted, the used environmental data are often based on data crawled by weather departments websites on the same day, and the real-time performance and the accuracy of the data have certain defects. The real-time property is mainly represented as follows: the granularity of data crawled by a meteorological department is coarse, and the data are often taken day by day, so that the prediction time range is relatively general; the accuracy is mainly reflected in: the predicted devices are often several devices within a region.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the invention provides a communication management system dynamic ring defect order assigning method, which is used for solving the technical problems of poor real-time performance and low accuracy of the existing power communication network fault information prediction method based on machine learning, and comprises the following steps: collecting the dynamic ring alarm data of the communication management system; inputting the collected dynamic ring alarm data into a trained defect fault prediction model, and outputting the defect fault information of the communication management system, wherein the fault prediction model is obtained by performing machine learning training on a deep residual error neural network; and generating the maintenance work order information to be distributed according to the defect fault information of the communication management system.
The embodiment of the invention also provides a dynamic ring defect dispatching device of a communication management system, which is used for solving the technical problems of poor real-time performance and low accuracy of the existing power communication network fault information prediction method based on machine learning, and comprises the following steps: the alarm data acquisition module is used for acquiring the dynamic ring alarm data of the communication management system; the fault prediction module is used for inputting the collected dynamic ring alarm data into a trained fault prediction model and outputting the fault information of the communication management system, wherein the fault prediction model is obtained by performing machine learning training on a deep residual error neural network; and the work order information generating module is used for generating maintenance work order information to be distributed according to the defect fault information of the communication management system.
The embodiment of the invention also provides computer equipment for solving the technical problems of poor real-time performance and low accuracy of the existing power communication network fault information prediction method based on machine learning.
The embodiment of the invention also provides a computer readable storage medium, which is used for solving the technical problems of poor real-time performance and low accuracy of the existing power communication network fault information prediction method based on machine learning.
In the embodiment of the invention, machine learning is carried out through a pre-depth residual error neural network, a defect fault prediction model is obtained through training, after the moving ring alarm data of a communication management system are collected, the collected moving ring alarm data are input into the defect fault prediction model, the defect fault information of the communication management system is output, and then the maintenance work order information to be distributed is generated according to the defect fault information of the communication management system. According to the embodiment of the invention, the defect fault prediction model is trained based on the deep residual error neural network, the defect fault can be predicted by utilizing large-scale dynamic ring alarm data, the prediction accuracy of the fault defect is greatly improved, and further, a corresponding maintenance work order is generated according to the predicted defect fault information, so that the practicability, intelligence and high efficiency of automatic dispatching work of the dynamic ring defect work order can be greatly improved, the operation and maintenance work of a communication management system is promoted, and the evolution is carried out towards automatic operation and maintenance and intelligent operation and maintenance.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
fig. 1 is a flowchart illustrating a method for assigning an order to a dynamic ring defect of a communication management system according to an embodiment of the present invention;
FIG. 2 is a flow chart of a model training method provided in an embodiment of the present invention;
FIG. 3 is a flow chart of a fault diagnosis knowledge base construction provided in an embodiment of the present invention;
fig. 4 is a schematic diagram of a dynamic ring defect dispatch device of a communication management system according to an embodiment of the present invention;
fig. 5 is a schematic diagram of an alternative dynamic ring defect order dispatching device of the communication management system according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a computer device provided in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
With the continuous application of new technologies in power communication operation and maintenance, the successful trial of technologies such as intelligent optical distribution (OASS) and optical cable online monitoring (OAM) not only promotes the automation of optical cable operation and maintenance, but also realizes the real-time grasp of the performance and the breakpoint of the optical cable fiber core. OASS and OAM make it possible to obtain real-time, on-the-spot optical cable performance and environmental data, have promoted the real-time nature and the comprehensiveness of operation and maintenance personnel to the control of electric power communication network.
In addition, the deep learning method and the general machine learning method are different mainly in that: the deep learning has good self-learning function and self-adaptive capacity, the deep learning has less human intervention in the characteristic engineering stage, the influence of artificial priori knowledge on the analysis result is reduced, and the external cause and the internal cause of the communication fault can be deeply mined.
Therefore, in the application of the automatic fault dispatching function of the communication management system, the deep learning algorithm is utilized to carry out fault early warning and diagnosis, the dynamic communication ring network management system is assisted to realize intelligent fault dispatching, the intelligent level of network operation and maintenance assistance can be improved, the practical process of the automatic fault dispatching function is promoted, the professional management and control efficiency is improved, and the professional operation and maintenance mode and the management mode of communication are safer and more reliable.
In view of the foregoing background, an embodiment of the present invention provides a dynamic ring defect order assignment method for a communication management system, and fig. 1 is a flowchart of the dynamic ring defect order assignment method for the communication management system, as shown in fig. 1, the method includes the following steps:
s101, collecting the moving loop alarm data of the communication management system.
It should be noted that the moving loop alarm data collected in S101 may be operation data or environment data of various devices in the communication management system monitored by the moving loop monitoring system. Because the scale of the dynamic ring alarm data is huge, the training of a shallow model or a traditional neural network for prediction can cause under-fitting, and therefore the depth of the model is very important. In order to meet the requirement of accurate prediction, in the embodiment of the invention, a deep residual error network in deep learning is utilized to train a fault defect prediction model.
And S102, inputting the collected dynamic ring alarm data into a trained defect fault prediction model, and outputting the defect fault information of the communication management system, wherein the fault prediction model is obtained by performing machine learning training on the deep residual error neural network.
It should be noted that the deep residual error neural network is a new variation of the convolutional neural network. Similar to the traditional convolutional neural network, the convolutional layer and the downsampling layer are also included, and the neurons of the convolutional layer are connected with the former layer in a local connection and weight sharing mode, so that the number of parameters needing to be trained is greatly reduced. The down-sampling layer can greatly reduce input dimensionality, reduce network complexity, enable the network to have higher robustness and effectively inhibit the over-fitting problem. As for the current academic understanding, if the network is deeper, the network performance will be better. Sometimes, however, deep neural networks have higher training and testing errors than shallow neural networks, and this phenomenon is called "degeneration". Furthermore, as the depth of the network increases, the gradient decreases in the backward direction, resulting in insufficient training of the convolutional layer in front of the network, which is called "gradient vanishing". The two problems lead to the difficulty in deep-level network structure training and learning, and the traditional convolutional neural network has only 20-30 layers.
The deep residual error network differs from the convolutional neural network in that it lets convolutional layers not directly learn the target, but instead learns a residual error, which is learned by building "identity maps" on different convolutional layers. Without considering the activation function, the neural network has a linear transformation between each layer, and the "identity mapping" can bypass the linear transformation, so that the network can be made not to degrade with the increase of depth by overlapping layers with y equal to x on the basis of a shallow network. This reflects the inability of multi-layer nonlinear networks to approach an identity mapping network.
And S103, generating maintenance work order information to be distributed according to the defect fault information of the communication management system.
In an embodiment, as shown in fig. 2, the method for assigning a defect in a dynamic ring of a communication management system according to an embodiment of the present invention further includes the following steps:
s201, acquiring training sample data and test sample data;
s202, according to training sample data, machine learning is carried out on a pre-constructed deep residual error neural network, and a defect fault prediction model is obtained through training;
and S203, testing the defect fault prediction model obtained by training according to the test sample data until a preset convergence condition is met.
It should be noted that the preset convergence condition may be that the accuracy of the model is higher than a preset accuracy or that the error of the model is lower than a preset error rate.
Because the network depth of the depth residual error network is far greater than that of the traditional convolutional neural network, the depth residual error network is easier to overfit, and therefore, in one embodiment, the communication management system dynamic ring defect order-assigning method provided by the embodiment of the invention can optimize the depth residual error neural network in the training process by using a Dropout regularization optimization algorithm. The use of the Dropout regularization optimization algorithm avoids overfitting of the neural network during training.
It should be noted that Dropout is a commonly used regularization method, which can suppress the occurrence of the over-fitting phenomenon to some extent and can improve the generalization capability of the network. The idea of Dropout is simple: in each training, the output value of a part of nodes is changed to 0 with a certain probability p, which is equivalent to that the part of nodes are 'deleted' from the whole network in the training, so that the corresponding parameters are not updated when the nodes are reversely propagated. And in the testing stage, the complete network is used for testing.
In order to prevent the overfitting, in an embodiment, the dynamic ring defect policy assignment method of the communication management system provided in the embodiment of the present invention may further construct a depth residual error neural network including a BN layer, where the BN layer is used to normalize data of each layer in the depth residual error neural network.
It should be noted that Batch Normalization (BN) can normalize each layer of data in the network. Different from a common normalization method, the BN layer is a network layer which can be learned and has parameters, and the parameters in the BN layer also need to be trained in the model training process. And a BN layer is added into the deep residual error neural network, so that overfitting can be prevented, and the training speed of the model can be greatly improved. Due to the fact that the data volume needing to be processed in the embodiment of the invention is large (large-scale dynamic ring alarm data), the rate of model convergence can be remarkably improved by using BN.
Fig. 3 is a flow chart of a fault diagnosis knowledge base construction provided in the embodiment of the present invention, and as shown in fig. 3, the embodiment of the present invention finally provides a fault defect early warning method for a communication management system based on a deep learning algorithm through steps of index system establishment, feature engineering analysis, model construction, and the like based on a BP neural network technology, so as to realize comprehensive evaluation and early warning in advance for communication network devices.
The BP (Back propagation) neural network is a neural network learning algorithm, and is a hierarchical neural network consisting of an input layer, an intermediate layer and an output layer, wherein the intermediate layer can be expanded into multiple layers. All the neurons in adjacent layers are connected, and all the neurons in each layer are not connected, and when a pair of learning modes are provided for the network, all the neurons obtain the input response of the network to generate the connection Weight (Weight). And then correcting the connection weights layer by layer from the output layer through the intermediate layers in a direction of reducing the error between the desired output and the actual output, and returning to the input layer. The process is repeatedly and alternately carried out until the global error of the network tends to a given minimum value, namely the learning process is completed. The BP neural network is a deep learning model for entry, and needs a great deal of tuning.
A Radial Basis Function (RBF) neural network is a three-layer feedforward network with a single hidden layer. Because the RBF network simulates the neural network structure of locally adjusting and mutually covering receiving domains (or called Receptive Field-received fields) in the human brain, the RBF network is a local approximation network, can approximate any continuous function with any precision, and is particularly suitable for solving the classification problem. The method is mainly used for image processing, voice recognition, time series prediction, radar origin positioning, medical diagnosis, error processing detection, mode recognition and the like. RBF networks are most used for classification, among which the most widespread is the pattern recognition problem.
The Convolutional Neural Network (CNN) is a hierarchical network, and only the function and form of the layer are changed, so that it is an improvement of the conventional neural network. At present, the method is mainly applied to image processing and has the advantages that: firstly, a convolution kernel is shared, and no pressure is applied to high-dimensional data processing; secondly, the characteristics do not need to be selected manually, the weight is trained well, and the characteristic classification effect is good. The disadvantages are that: firstly, parameter adjustment is needed, a large sample amount is needed, and the best GPU is needed for training; ② the physical meaning is ambiguous (i.e. we do not know what features each convolutional layer extracts at all, and the neural network itself is a "black box model" that is difficult to interpret).
A Recurrent Neural Network (RNN) is a neural network with a feedback structure whose outputs are related not only to the current inputs and the weights of the network, but also to the inputs of the previous network, the RNN modeling time by adding a self-connecting hidden layer that spans the time points. RNNs mainly process time series data, for example, words are sequential for a certain sentence, so that context understanding can be better performed, and therefore, the RNNs are often used in the field of NLP, such as machine translation, emotion analysis, and the like.
And (4) optimizing different models through repeated experiments, and finally determining that the effect of the defect fault prediction model obtained based on deep residual error network training is the best.
Based on the same inventive concept, the embodiment of the present invention further provides a dynamic ring defect dispatching device of a communication management system, as described in the following embodiments. Because the principle of solving the problems of the device is similar to the method for assigning the list according to the dynamic ring defect of the communication management system, the implementation of the device can refer to the implementation of the method for assigning the list according to the dynamic ring defect of the communication management system, and repeated parts are not described again.
Fig. 4 is a schematic diagram of a dynamic ring defect dispatch apparatus of a communication management system according to an embodiment of the present invention, as shown in fig. 4, the apparatus includes: an alarm data acquisition module 41, a defect fault prediction module 42 and a work order information generation module 43.
The alarm data acquisition module 41 is configured to acquire moving loop alarm data of the communication management system; the defect fault prediction module 42 is configured to input the collected dynamic ring alarm data into a trained defect fault prediction model, and output defect fault information of the communication management system, where the defect fault prediction model is obtained by performing machine learning training on a deep residual error neural network; and a work order information generating module 43, configured to generate repair work order information to be distributed according to the defect fault information of the communication management system.
In an embodiment, as shown in fig. 5, the apparatus for assigning a defect on a dynamic ring of a communication management system according to an embodiment of the present invention further includes: a sample data obtaining module 44, configured to obtain training sample data and test sample data; the model training module 45 is used for performing machine learning on the pre-constructed deep residual error neural network according to training sample data, and training to obtain a defect fault prediction model; and the model testing module 46 is configured to test the trained defect fault prediction model according to the test sample data until a preset convergence condition is met.
In an embodiment, in the dynamic loop defect assignment apparatus for a communication management system provided in the embodiment of the present invention, the model training module 45 is further configured to optimize the deep residual neural network in the training process by using a Dropout regularization optimization algorithm.
In an embodiment, as shown in fig. 5, the apparatus for assigning a defect on a dynamic ring of a communication management system according to an embodiment of the present invention further includes: and a depth residual neural network constructing module 47, configured to construct a depth residual neural network including a BN layer, where the BN layer is configured to perform normalization processing on data of each layer in the depth residual neural network.
Based on the same inventive concept, a computer device is further provided in the embodiments of the present invention to solve the technical problems of poor real-time performance and low accuracy of the existing power communication network fault information prediction method based on machine learning, fig. 6 is a schematic diagram of a computer device provided in the embodiments of the present invention, as shown in fig. 6, the computer device 60 includes a memory 601, a processor 602, and a computer program stored in the memory 601 and operable on the processor 602, and the processor 602 implements the above-mentioned dynamic ring defect order assignment method of the communication management system when executing the computer program.
Based on the same inventive concept, the embodiment of the invention also provides a computer readable storage medium, which is used for solving the technical problems of poor real-time performance and low accuracy of the existing power communication network fault information prediction method based on machine learning.
In summary, embodiments of the present invention provide a method, an apparatus, a computer device, and a computer-readable storage medium for assigning a dynamic ring defect of a communication management system, where machine learning is performed through a deep residual error neural network in advance, a defect fault prediction model is obtained through training, after dynamic ring alarm data of the communication management system is collected, the collected dynamic ring alarm data is input into the defect fault prediction model, defect fault information of the communication management system is output, and then repair work order information to be assigned is generated according to the defect fault information of the communication management system. According to the embodiment of the invention, the defect fault prediction model is trained based on the deep residual error neural network, the defect fault can be predicted by utilizing large-scale dynamic ring alarm data, the prediction accuracy of the fault defect is greatly improved, and further, a corresponding maintenance work order is generated according to the predicted defect fault information, so that the practicability, intelligence and high efficiency of automatic dispatching work of the dynamic ring defect work order can be greatly improved, the operation and maintenance work of a communication management system is promoted, and the evolution is carried out towards automatic operation and maintenance and intelligent operation and maintenance.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (10)

1. A dynamic ring defect order dispatching method for a communication management system is characterized by comprising the following steps:
collecting the dynamic ring alarm data of the communication management system;
inputting the collected dynamic ring alarm data into a trained defect fault prediction model, and outputting the defect fault information of the communication management system, wherein the fault prediction model is obtained by performing machine learning training on a deep residual error neural network;
and generating the maintenance work order information to be distributed according to the defect fault information of the communication management system.
2. The method of claim 1, wherein the method further comprises:
acquiring training sample data and test sample data;
according to the training sample data, machine learning is carried out on a pre-constructed deep residual error neural network, and a defect fault prediction model is obtained through training;
and testing the defect fault prediction model obtained by training according to the test sample data until a preset convergence condition is met.
3. The method of claim 2, wherein the deep residual neural network in the training process is optimized using a Dropout regularization optimization algorithm.
4. The method of claim 2, wherein the method further comprises:
and constructing a depth residual error neural network comprising a BN layer, wherein the BN layer is used for carrying out normalization processing on data of each layer in the depth residual error neural network.
5. A communication management system dynamic ring defect order dispatching device is characterized by comprising:
the alarm data acquisition module is used for acquiring the dynamic ring alarm data of the communication management system;
the fault prediction module is used for inputting the collected dynamic ring alarm data into a trained fault prediction model and outputting the fault information of the communication management system, wherein the fault prediction model is obtained by performing machine learning training on a deep residual error neural network;
and the work order information generating module is used for generating maintenance work order information to be distributed according to the defect fault information of the communication management system.
6. The apparatus of claim 5, wherein the apparatus further comprises:
the sample data acquisition module is used for acquiring training sample data and test sample data;
the model training module is used for performing machine learning on a pre-constructed deep residual error neural network according to the training sample data, and training to obtain a defect fault prediction model;
and the model testing module is used for testing the defect fault prediction model obtained by training according to the test sample data until a preset convergence condition is met.
7. The apparatus of claim 6, in which the model training module is further to optimize a deep residual neural network in a training process using a Dropout regularization optimization algorithm.
8. The apparatus of claim 6, wherein the apparatus further comprises:
the depth residual error neural network construction module is used for constructing a depth residual error neural network comprising a BN layer, wherein the BN layer is used for carrying out normalization processing on data of each layer in the depth residual error neural network.
9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the communication management system dynamic ring defect policy assignment method according to any one of claims 1 to 4 when executing the computer program.
10. A computer-readable storage medium storing a computer program for executing the method of dynamic ring defect dispatch for a communication management system according to any of claims 1 to 4.
CN202110254521.XA 2021-03-09 2021-03-09 Method and device for assigning orders of dynamic ring defects of communication management system Pending CN112862211A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110254521.XA CN112862211A (en) 2021-03-09 2021-03-09 Method and device for assigning orders of dynamic ring defects of communication management system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110254521.XA CN112862211A (en) 2021-03-09 2021-03-09 Method and device for assigning orders of dynamic ring defects of communication management system

Publications (1)

Publication Number Publication Date
CN112862211A true CN112862211A (en) 2021-05-28

Family

ID=75993614

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110254521.XA Pending CN112862211A (en) 2021-03-09 2021-03-09 Method and device for assigning orders of dynamic ring defects of communication management system

Country Status (1)

Country Link
CN (1) CN112862211A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113743512A (en) * 2021-09-07 2021-12-03 上海观安信息技术股份有限公司 Autonomous learning judgment method and system for safety alarm event

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110031226A (en) * 2019-04-12 2019-07-19 佛山科学技术学院 A kind of diagnostic method and device of bearing fault
CN111830408A (en) * 2020-06-23 2020-10-27 朗斯顿科技(北京)有限公司 Motor fault diagnosis system and method based on edge calculation and deep learning
CN112171721A (en) * 2020-11-30 2021-01-05 北京科技大学 Robot joint sensor and actuator fault diagnosis method and system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110031226A (en) * 2019-04-12 2019-07-19 佛山科学技术学院 A kind of diagnostic method and device of bearing fault
CN111830408A (en) * 2020-06-23 2020-10-27 朗斯顿科技(北京)有限公司 Motor fault diagnosis system and method based on edge calculation and deep learning
CN112171721A (en) * 2020-11-30 2021-01-05 北京科技大学 Robot joint sensor and actuator fault diagnosis method and system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113743512A (en) * 2021-09-07 2021-12-03 上海观安信息技术股份有限公司 Autonomous learning judgment method and system for safety alarm event

Similar Documents

Publication Publication Date Title
CN111461463B (en) Short-term load prediction method, system and equipment based on TCN-BP
Lechevalier et al. A neural network meta-model and its application for manufacturing
CN116245033B (en) Artificial intelligent driven power system analysis method and intelligent software platform
CN116467674B (en) Intelligent fault processing fusion updating system and method for power distribution network
CN112733440A (en) Intelligent fault diagnosis method, system, storage medium and equipment for offshore oil-gas-water well
Chen et al. A multivariate grey RBF hybrid model for residual useful life prediction of industrial equipment based on state data
Chen et al. Transfer life prediction of gears by cross-domain health indicator construction and multi-hierarchical long-term memory augmented network
CN116523187A (en) Engineering progress monitoring method and system based on BIM
CN112434813A (en) Multiple multidimensional language reasoning method based on attribute-oriented language concept lattice
Wang et al. Reliability analysis of complex electromechanical systems: State of the art, challenges, and prospects
CN112862211A (en) Method and device for assigning orders of dynamic ring defects of communication management system
Uljayev et al. Application of expert systems for measuring the humidity of bulk materials
CN106845144A (en) A kind of trend prediction method excavated based on industrial big data
CN117154680A (en) Wind power prediction method based on non-stationary transducer model
CN115983714A (en) Static security assessment method and system for edge graph neural network power system
Azyus et al. Determining rul predictive maintenance on aircraft engines using gru
Zhao et al. Knowledge-Informed Uncertainty-Aware Machine Learning for Time Series Forecasting of Dynamical Engineered Systems
CN112510699A (en) Transformer substation secondary equipment state analysis method and device based on big data
Takács et al. The AHP extended fuzzy based risk management
Najar et al. Comparative Machine Learning Study for Estimating Peak Cladding Temperature in AP1000 Under LOFW
CN111582632B (en) Multi-factor safety stage prediction method for whole process of underground large space construction
Misnik et al. Technique for development of information-analytical processes in cyber-physical systems based on neural-fuzzy petri nets
Su et al. A Three-Dimensional Information Flow Modeling Method for Integrated Fault Diagnosis and Maintenance of Complex System
Shchehlov et al. Availability Markov Models of Complex Equipment with Relearning Diagnostic Means after Detecting Latent Failures
Jin et al. Overview of Research on Health Assessment and Fault Prediction of Complex Equipment Driven by Big Data

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