Unauthorized activities remain a leading cause of pipeline incidents. High Fidelity Distributed Sensing (HDS) is an optical sensing solution that has proven particularly effective in monitoring applications including leak detection, strain monitoring, and pig tracking. Due to its high sensitivity and real time nature, HDS is also very well-suited for the detection of intrusion events like unauthorized digging or right-of-way (ROW) crossings. Successful implementation of intrusion detection requires more than just system deployment - it demands a practical, collaborative approach that integrates technology with operational workflows.

A case study will be presented of how a pipeline operator and Hifi partnered to implement HDS for intrusion detection on a massive scale – monitoring every meter of a pipeline for over one thousand kilometers. Practical steps needed to be taken to ensure the system delivered actionable intelligence without overwhelming operators with false or nuisance alarms.

Targeted field intrusion simulations performed at various distances from the pipeline ROW guided the creation of the intrusion detection machine learning (ML) models, followed by a controlled rollout to select areas to validate performance and minimize alarm fatigue.

Most pipelines are in proximity of construction activity, which can trigger alarms even if not posing a direct threat. Weekly look-ahead notifications of planned activity were incorporated into the monitoring process to suppress alarms generated in permitted areas. A web form was developed to allow the operator to input planned activity, enabling automated suppression of alarms in the algorithms.

The detection of real-world incidents has demonstrated great value. These notifications have enabled the operator’s integrity teams to assess potential threats – via vehicle crossing calculations, for example - and take proactive measures where appropriate.

Lessons learned will be included on the importance of human factors such as communication and operating philosophies in maximizing the value of the HDS monitoring.