导弹飞行试验数据预处理方法

A preprocessing method of missile flight test data

在导弹的挂飞和靶试等飞行试验中,一般涉及载机、目标机/靶机、导弹、地面测量站等多方面的试验主体和试验测量记录单元。试验结束后,为了快速的从各方数据中获得有效的信息、准确的评估试验是否成功等,均需高效的对所有飞行相关的数据进行预处理,以获得有效的、平滑的、时间统一的、融合后的数据。文章首先研究了数据预处理中的数据挖掘、噪声处理、缺失数据处理、时钟统一等关键技术,在噪声处理、缺失数据处理和时钟统一方面结合实际工程经验提出了高效实用的数据预处理技术,最后介绍了某项飞行试验中所开发的数据预处理软件及实现过程。经过软件开发及测试、使用,验证了关键技术使用和处理的正确性、可行性。

In missile flight tests, the recording units of carrier aircraft,target aircraft, missile, ground measurement station and so on are generally involved. After the test, in order to quickly obtain effective information from the data and accurately evaluate the success of the test, it is necessary to efficiently preprocess all flight related data. Firstly, this paper studies the key technologies in data preprocessing, such as data mining, noise processing, missing data processing and clock unification. In terms of noise processing, missing data processing and clock unification, combined with practical engineering experience, this paper puts forward efficient and practical data preprocessing technology. Finally, it introduces the data preprocessing software developed in a flight test and its implementation process. After application, the correctness and feasibility of the use of key technologies are verified.