Font Size: a A A

Design And Implementation Of Generalized Data Recording And Analysis Software

Posted on:2021-05-01Degree:MasterType:Thesis
Country:ChinaCandidate:S WangFull Text:PDF
GTID:2492306512992179Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
In the daily testing of complex weapon systems,the performance of the system needs to be evaluated based on various types of data in the system.First,recording all types of data in the weapon system and parsing them according to the communication protocol.And evaluate the system performance based on the analyzed data or diagnose the fault of the weapon system by using the fault parameter data to help the tester locate the fault quickly,which is conducive to the improvement and maintenance of the weapon systems.Thinking of data processing,the thesis designs and completes the data recording and analysis,simulates the diagnosis and prediction of missile fault and uses the simulation results to complete the fault diagnosis.The main work completed in this thesis includes the following aspects:(1)The communication data between weapon systems are recorded based on the Win Pcap library.Users can set filter rules and filter conditions to record various types of data packets and view historical data.Setting filter rules and conditions through the interactive interface.They can also parse the recorded data by using external XML documents instead of the traditional way of describing protocols in software code.Improving the generalization of software.(2)Using the parsed data for fault diagnosis and prediction.Based on the historical data of missile faults,a fault tree is established.Relevant parameters affected when a fault tree event occurs are cleared,and parameter data of various modes when the fault occurs are used as a training set,and the fault is diagnosed using BP neural network.According to the time series of the fault parameters,the fault prediction is performed to determined whether a fault will occur at the next time period.(3)The optimal solution obtained by training the BP neural network is used as the calculation parameter of the fault diagnosis module to implement the fault diagnosis module.Diagnose the fault parameters recorded during the test,quickly locate the fault location in combination with the established fault tree,and record the fault in the database.You can evaluate the performance of a component based on the frequency of component failures.(4)Testing and verifying the performance and functions of the data recording module,analysis module and fault diagnosis module,and improve the defects found during the verification process.
Keywords/Search Tags:Data logging, Database, Data analysis, WinPcap, XML, Fault diagnosis, Fault prediction
PDF Full Text Request
Related items