基于RS_Hash频繁项集的卫星载荷关联规则算法

Algorithm of satellite payload association rules based on RS_Hash frequent item sets

遥测数据是反映卫星健康状态的重要依据,对遥测载荷数据进行关联性分析,在一定程度上能反映出卫星的整体运行情况的好坏。针对传统关联规则算法存在效率低下、占用内存过多的问题,提出一种基于RS_Hash频繁项集的卫星载荷关联规则算法。首先对事务数据库使用动态随机抽样的方法获取样本数据,设计抽样误差和抽样停止规则来确定最优的样本容量;其次将抽取出的样本使用哈希桶来存储频繁项集,进而减少占用的内存,提高算法的运行效率;最后使用3个与载荷数据相似的公开数据集和卫星载荷数据集进行实验,结果表明,在公共数据集上取得了良好的效果,尤其是在具有大数据量级的卫星载荷数据集上效果明显,在不同事务长度和支持度的情况下,相较于Apriori、PCY、SON、FP-Growth、RCM_Apriori和Hash_Cumulate算法,RS_Hash算法在平均时间效率上分别提高了75.81%、49.10%、59.38%、50.22%、40.16%和39.22%。

Telemetry data is an important basis to reflect the health status of satellites.The correlation analysis of telemetry load data can reflect the overall operation status of satellites to a certain extent.Aiming at the problems of low efficiency and excessive memory consumption of traditional association rule algorithm,a satellite load association rule algorithm based on RS_Hash frequent item set was proposed.Firstly,the dynamic random sampling method is used to obtain the sample data of the transaction database,and the sampling error and sampling stop rules are designed to determine the optimal sample size.Secondly,hash buckets are used to store frequent item sets in the extracted samples,thus reducing the occupied memory and improving the operation efficiency of the algorithm.Finally,three public data sets which are similar to the load data and satellite load data sets are used to carry out experiments.The results show that good results are achieved on the public data sets,especially on the satellite load data sets with large data magnitude.Under the condition of different transaction lengths and support degrees,compared with Apriori,PCY,SON,FPGrowth,RCM_Apriori and Hash_Cumulate algorithm,the average time efficiency of RS_Hash algorithm is improved by 75.81%,49.10%,59.38%,50.22%,40.16%and 39.22%.