面向状态评估的装备测试数据处理

Equipment Test Data Processing Aiming at Condition Assessment

从测试数据分类出发,主要针对模拟量参数进行数据处理,对数据进行清洗、合成与整理,解决了数据冗余问题;对数据进行归一化处理,解决了数据种类、量纲不统一问题;运用改进的时间序列算法对缺失数据进行填充,解决了测试数据不完备问题;采用主成分分析法提取状态评估关键参数,解决了参数众多、测试次数累积造成的维数灾难问题。算例证明,经过处理后的装备测试数据更加完备规范,可有效应用于装备状态评估。

This article embarks from the test data classification.The data processing is mainly for analog parameters while the data is cleaned,synthetized,processed to solve the problem of data redundancy and then the data is normalized to unify the data types and dimensions.After this the improved algorithm of time series is used to fill the missing data for data completion.Principal component analysis(PCA)is further served as method to extract key parameters for condition assessment to solve tests Cumulative Dimension Disaster of huge number of parameters.The example shows that the processed equipment test data is more complete and normative,which can be effectively applied to the equipment condition assessment.