Electric propulsion reliability: statistical analysis of on-orbit anomalies and comparative analysis of electric versus chemical propulsion failure rates

摘要

With a few hundred spacecraft launched to date with electric propulsion (EP),it is possible to conduct an epidemiological study of EP on orbit reliability.The first objective of the present work was to undertake such a study andanalyze EP track record of on orbit anomalies and failures by differentcovariates. The second objective was to provide a comparative analysis of EPfailure rates with those of chemical propulsion. After a thorough datacollection, 162 EP-equipped satellites launched between January 1997 andDecember 2015 were included in our dataset for analysis. Several statisticalanalyses were conducted, at the aggregate level and then with the datastratified by severity of the anomaly, by orbit type, and by EP technology.Mean Time To Anomaly (MTTA) and the distribution of the time to anomaly wereinvestigated, as well as anomaly rates. The important findings in this workinclude the following: (1) Post-2005, EP reliability has outperformed that ofchemical propulsion; (2) Hall thrusters have robustly outperformed chemicalpropulsion, and they maintain a small but shrinking reliability advantage overgridded ion engines. Other results were also provided, for example thedifferentials in MTTA of minor and major anomalies for gridded ion engines andHall thrusters. It was shown that: (3) Hall thrusters exhibit minor anomaliesvery early on orbit, which might be indicative of infant anomalies, and thuswould benefit from better ground testing and acceptance procedures; (4) Strongevidence exists that EP anomalies (onset and likelihood) and orbit type aredependent, a dependence likely mediated by either the space environment ordifferences in thrusters duty cycles; (5) Gridded ion thrusters exhibit bothinfant and wear-out failures, and thus would benefit from a reliability growthprogram that addresses both these types of problems.