key: cord-0061024-4v9ex79t
authors: Jia, Shuang-cheng; Wang, Tao
title: Research on Fault Intelligent Detection Technology of Dynamic Knowledge Network Learning System
date: 2020-06-08
journal: Multimedia Technology and Enhanced Learning
DOI: 10.1007/978-3-030-51100-5_39
sha: d2cd5a0a9e9676a97f2f4634f6f7b9ce8e54c644
doc_id: 61024
cord_uid: 4v9ex79t

The rapid development of computers has improved the scope of dynamic knowledge network learning applications. Online learning has brought convenience to people in time and place. At the same time, people began to pay attention to the efficiency and quality of online learning. At present, the network knowledge storage system is distributed storage system. The distributed storage system has great performance in terms of capacity, scalability, and parallelism. However, its storage node is inexpensive, and the reliability is not high, and it is prone to fault. Based on the designed fault detection model detection path, relying on building the knowledge data node fault detection mode, constructing the knowledge data link fault detection mode, completing the fault detection model detection mode, and finally realizing the dynamic knowledge network learning system fault intelligent detection technology research. The experiment proves that the dynamic knowledge network learning system fault intelligent detection technology designed in this paper reduces the fault rate of the network learning system by 37.5%.

Network learning advocates students as the main body, while intelligent network learning system is characterized by students' personalized learning [1] . The organization of domain knowledge and intelligent navigation are the basic problems to realize personalized learning in intelligent network learning system [2] . In this era of information explosion, learning determines its own competitiveness. However, many people spend a lot of time and money to enrich themselves through various channels. As a result, they find that what they finally master is fragmented information, which can not form a complete knowledge framework [3] . These isolated knowledge can not be used to solve practical problems, so it is particularly important to build a perfect knowledge learning system.

The dynamic knowledge network learning system jointly created by network technology, computer technology and communication technology has become an emerging learning method, providing the masses with convenient conditions for learning anywhere and anytime [4] . The storage system in the network learning system is composed of multiple storage points, but multiple storage points are prone to data synchronization failure and serial effects [5] . When a virus invades one of the storage points, the relevant storage points may be implicated. Based on the above situation, the network learning system should adopt the fault intelligent detection technology to automatically detect the entire learning system before the network knowledge system fails, and find the place where the fault may occur, by backing up, copying, transferring, etc. the knowledge data. Which means to process the network learning system to ensure the integrity of the knowledge data and the security of the network learning system [6] . The experiment proves that the network learning system using the automatic fault monitoring technology can greatly reduce the fault rate of the system itself. In order to reduce the failure rate of dynamic knowledge network learning system, this paper proposes a new fault detection technology for dynamic knowledge network learning system.

The most common faults in dynamic knowledge network learning systems are data loss and data confusion. The system loses a lot of data and the system knowledge cannot be acquired normally. Set the data loss rate of the dynamic knowledge network learning system as u k ¼ Kx k , the following formula can be obtained.

Assuming the probability of data confusion is r, then:

If the estimated error of the state observation parameters of the dynamic knowledge network learning system is set as eðkÞ ¼ xðkÞ À b xðkÞ, no data loss or confusion will occur. The state estimation error can be calculated by using the following formula:

In the above formula, the following formula can be obtained by introducing the

The following formula is used to represent the estimation error of state observation parameters in the case that data of dynamic knowledge network learning system cannot be obtained.

The fault detection path model of random switching control can be established by using the following formula:

The following state observation parameters can make the fault path detection model approach stable.

In the above formula, KðkÞ is used to describe the kalman filter increment. According to the method described above, the fault detection path of dynamic knowledge network learning system is designed by using the residual of observation parameters. The detection path is shown in Fig. 1 .

Other knowledge systems The fault detection path ensures the security of the knowledge data in a single disk by copying and transferring [7, 8] . If the knowledge data in the storage system is lost, all knowledge acquisition in the entire network learning system cannot be performed normally [9] . Therefore, ensuring the security of knowledge data in the system is a top priority.

In the dynamic knowledge network learning system, each operation execution time and system feedback time delay of acquiring knowledge are negligible [10] . The network learning system applies network technology to transmit knowledge information, and at the same time, the system has the possibility of fault [11] .

Let G = (p, K) represent the network in the knowledge learning system, G is a vector, and n = (1,…, n) represents the set of knowledge storage points in the knowledge learning system [12] . Then, K p  p represents a collection of knowledge links in the knowledge learning system. Expressed as function [13] :

Where F Y is the value range of the knowledge system, F X is the natural number set, and formula (1) represents the set of knowledge nodes. A indicates the set of storage node faults at a certain moment. If the input data produces set A, it indicates that the knowledge data node is not faulty, and the knowledge data is not lost in the storage system [14] . If the input data does not generate the set A, it indicates that the knowledge data node is faulty, and the corresponding knowledge data is lost. The dynamic knowledge network learning system will automatically detect the cause of the loss and synchronize the data in time.

The link refers to the path passed by the knowledge node in the network knowledge system. The two knowledge nodes belong to the neighbor relationship, and the network path from the m knowledge node to the n knowledge node is represented by m-n, and the two adjacent relationship knowledge nodes in the knowledge system [15] , using neighbor(n) to represent its path set. There are two functions for the link between the node m and the n node, namely: send m,n (A) and receive m,n (A). The former is to pass the information A in the knowledge node m to the knowledge node n, if the knowledge node n receives the information A in the knowledge node m, then the knowledge node m passes the latter function to the knowledge point n [16] . Expressed by function:

Where y m is the information amount of the m node, T d is the difference between the information amounts of the m node and the n node, and R is a natural integer. For the knowledge storage nodes m and n, under the premise that the m-n link is complete, the knowledge storage node m transmits information to the n through the link. If the send m, n (A) and receive m,n (A) functions are displayed in the system, it means that the link is not faulty, and the information won't be lost while transmitted in the link. If both send m,n (A) and receive m,n (A) functions are not displayed in the system, or only one [17] is displayed, it indicates that there is a problem between the two knowledge nodes, the dynamic knowledge network learning system will automatically monitor the cause of the problem and fix it in time.

When a knowledge node or a link in the dynamic knowledge network learning system has problems, it will affect the normal operation of the knowledge learning system [18, 19] . The introduction of fault self-energy monitoring technology is mainly to automatically detect the entire learning system and find out where the fault may occur before the network knowledge system fails [20] . Before the fault occurs, the network learning system is processed by means of backing up, copying, and transferring the knowledge data to ensure the integrity of the knowledge data and the security of the network learning system [21, 22] . The fault detection algorithm is as follows:

Var Dp: //Fault detection module of initial process p;

Rp "π": The initial process p has a corresponding timer for any of the processes in the set π;

Dp=φ;

Forall q do rp "q=δ+υ";

Start sending process,send message with interval δ:

Forall q do send <alive> to q;

Receive process, q reset after receiving <alive> message: Rp "q=δ+υ";

Detecte faults when rp "q" exceeds interval δ:

The fault detection technology is based on the delay of knowledge node information [23] , d is the transmission period of information fault detection, and the knowledge data node sends the <alive> message [24, 25] according to the d period, if the response feedback message has not been received within a certain time, which indicates that the dynamic knowledge network learning system has failed (Fig. 2) .

Based on the fault detection model detection path design, the knowledge data node fault detection mode is built, the knowledge data link fault detection mode is built, and the network learning system fault intelligent detection model is realized.

In order to verify the effectiveness of the research on the fault intelligent detection technology of the dynamic knowledge network learning system proposed in this paper, the simulation test is carried out. During the test, two different dynamic knowledge network learning systems are used as test objects to test whether the system which gained knowledge per unit time failed. The knowledge data of two learning systems, English, mathematics, and physics are simulated. In order to ensure the accuracy of the simulation test, multiple simulation tests are performed, and the data generated by multiple tests are presented in the same data chart. 

In order to ensure the accuracy of the simulation test process and results, the test parameters are set. The simulation test uses two different dynamic knowledge network learning systems as the simulation test objects, simulates the number of system faults, and analyzes the simulation test results. Due to the amount of knowledge, the complexity of knowledge information, and the difficulty level of knowledge information contained in different stages and levels of knowledge data are different, the amount of correct knowledge is also different, and it has an impact on the test results. Therefore, it is necessary to ensure that the difficulty, subject and study time are consistent during the test. Set the tester's network learning time to 24 h. The test data setting results in this paper are shown in Table 1 .

The experimental results are counted as a contract collection schedule, and the test results are shown in Table 2 . The numbers behind the subjects in the table indicate the difficulty of knowledge. The larger the number, the more complicated and difficult the relevant knowledge is. Because 12 different execution parameters are set during the test, it is impossible to obtain the experimental results at a glance, thus affecting the comparative analysis of the two systems. According to Table 2 , the comparison of the number of faults of the system when acquiring knowledge is shown in Fig. 3 . In the figure, the abscissa is the degree of difficulty of knowledge, and the ordinate is the number of times the system fails when acquiring knowledge. The horizontal and vertical coordinate data are referred to Table 2 . It can be seen from Fig. 3 that the dynamic knowledge network learning system fault intelligent detection technology research is significantly reduced compared with the traditional system.

In order to further verify the superiority of this method, the failure rate of dynamic knowledge network learning system is compared after the application of different methods. The results are shown in Table 3 .

As can be seen from the above table, the intelligent fault detection technology of dynamic knowledge network learning system designed in this paper reduces the failure rate of network learning system by 37.5%. 

This paper proposes a research on intelligent detection technology for dynamic knowledge network learning system fault, based on the design of fault detection model detection path, relying on building knowledge data node fault detection mode, constructing knowledge data link fault detection mode, the construction of fault detection model detection mode is completed, and finally the fault intelligent detection technology studied in this paper is realized. The tests in this paper show that the research on fault intelligent detection technology is effective. It is hoped that the research of fault intelligent detection technology in this paper can provide a theoretical basis for the research of fault intelligent detection technology in dynamic knowledge network learning system.

Research on extraction and detection of intrusion feature information in mobile network

Research on a knowledge modelling methodology for fault diagnosis of machine tools based on formal semantics

Intelligent multi-area power control: dynamic knowledge domain inference concept

Intelligent information processing for building university knowledge base

Detection of driving events using sensory data on smartphone

A novel method for intelligent EMC management using a "knowledge base

Research on fault detection method for heat pump air conditioning system under cold weather

Mining students activities from a computer supported collaborative learning system based on peer to peer network. Peer-to-Peer Netw

Prediction of cascading collapse occurrence due to the effect of hidden failure of a protection system using artificial neural network

Cascading failure of command information system bi-layer coupled network model. Harbin Gongye Daxue Xuebao

Fault detection and diagnosis for building cooling system with a tree-structured learning method

Image fusion of fault detection in power system based on deep learning

A neural network-based intelligent cognitive state recognizer for confidence-based e-learning system

An integrated learning and filtering approach for fault diagnosis of a class of nonlinear dynamical systems

Eunomos, a legal document and knowledge management system for the Web to provide relevant, reliable and up-to-date information on the law

The tourism knowledge system

Exploring the logic and landscape of the knowledge system: multilevel structures, each multiscaled with complexity at the mesoscale. Engineering

Evolution of the knowledge system for agricultural development in the Yaqui Valley

The Internet of Things: building a knowledge management system for open innovation and knowledge management capacity

Toward a learning health-care systemknowledge delivery at the point of care empowered by big data and NLP

The Land-Potential Knowledge System (LandPKS): mobile apps and collaboration for optimizing climate change investments

Enhancing knowledge sharing intention through the satisfactory context of continual service of knowledge management system

Development of the knowledge system based on formation of holistic competence of project managers in the Czech Republic

Towards a responsible early-warning system: Knowledge implications in decision support design

Construction of expert knowledge monitoring and assessment system based on integral method of knowledge evaluation