The TUBE algorithm: Discovering trends in time series for the early detection of fuel leaks from underground storage tanks

作者：

Highlights：

• A new method for detecting trends in time series is proposed.

• Application of the algorithm to fuel leaks detection problem is presented.

• A step-by-step usage example, as well as time and memory complexity is presented.

• The tests have been performed in accordance to the American certification standard.

• The probability of detection equals 93.11% with 0.73% probability of false alarm.

摘要

•A new method for detecting trends in time series is proposed.•Application of the algorithm to fuel leaks detection problem is presented.•A step-by-step usage example, as well as time and memory complexity is presented.•The tests have been performed in accordance to the American certification standard.•The probability of detection equals 93.11% with 0.73% probability of false alarm.

论文关键词：Trend detection,Leak detection,Anomaly detection,Time series,Petrol station,Quality of data

论文评审过程：Received 1 March 2017, Revised 7 August 2017, Accepted 8 August 2017, Available online 14 August 2017, Version of Record 24 August 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.08.016