Research And Implementation Of Measurement Data Evaluation System For Aviation Industry

Measurement processes are extremely common in tests and productions in the aviation industry.As the aviation industry develops in the direction of high precision,the quality of measurement data is becoming more and more demanding.An accurate and scientific measurement evaluation can provide an important reference for the optimization of measurement strategy and work,thus improving the research and work efficiency of the aviation industry.At present,the evaluation processes of measurement data are complicated and tedious,and often take the rough manual calculations,which would lead to the low reliability of the results.Based on this,this thesis proposes to build a measurement data evaluation system for the aviation industry,and clarifies the software functions and technical methods required to realize the system.This thesis starts from the background and significance of building the evaluation system,then clarifies the research objectives and tasks of measurement evaluation.After that,from the perspective of software engineering,this thesis presents the complete construction plan of the evaluation system from the processes of demand analysis,toplevel design,software implementation and system test.This thesis also proposes some functions applicable to measurement data evaluation,such as pre-measurement analysis,post-measurement analysis,expert model and modeling simulation,and offers technical methods such as data pre-processing,data feature reconstruction,automatic uncertainty analysis,correlation impact analysis,scenario modeling simulation and etc.1.A method for accurate characterization of measurement data is proposed.Based on the different structural properties of the measurement data,a rule base and an improved isolation forest algorithm are used for data pre-processing to facilitate subsequent computational analysis.Through the feature test and reconstruction technical study,estimation system can accurately represent the measurement data with about 20 feature distribution models and grasp the mechanism of measurement data.2.Standard quantification of the measurement quality is performed.The system introduces the internationally accepted uncertainty method and realizes automatic quality evaluation of measurement parameters under both GUM and MCM methods by studying the four major processes of quantification,propagation,synthesis and distribution to accurately quantify the measurement quality.Also,the role of correlation impact analysis is to express the relationship between the components and the total from statistical principles such as ANOVA and mutual information.The results are reported using data visualization techniques,while assessment reports can be generated automatically.3.The modeling and simulating study of the measurement scenario is completed.By profoundly understanding the underlying physical principles of measurement scenes,the thesis studies measurement scene simulation and data trend prediction techniques,abstracts measurement layouts and constructs digital models of measurement components,finally forms virtual simulation systems corresponding to real physical measurements.This study enables the prediction of data trends after parameter condition changes and reveals the hidden development pattern of current measurement data.4.A software platform for measurement data evaluation is established.The front and back platform of the system are realized by using software development technology.This study units also builds a simple and easy-to-use graphical interface,effectively develops various modules,and realizes the effective organization and mutual collaboration of service modules which effectively support the measurement data evaluation works of aviation industry.