Flowstate co-founder and CEO Jerad Stack recently joined host Russell Treat on the Pipeliners Podcast for Episode 451, “Leak Detection is Not Just One Problem,” to explore how pipeline operators can think differently about computational pipeline monitoring (CPM) and the growing role of machine learning in leak detection.
During the conversation, Jerad and Russell discuss how different pipeline systems create fundamentally different leak detection challenges. A transmission line, a complex gathering system, and a pipeline operating in slack-line conditions each behave differently—and the technology monitoring them needs to account for that operational reality.
They also explore how Flowstate approaches the problem by using pipeline data already being collected—including flow, pressure, and temperature—to understand expected system behavior and identify meaningful anomalies.
What You’ll Hear in the Episode
The conversation covers several important questions facing pipeline operators today:
- Why does pipeline leak detection mean different things depending on the system and operating environment?
- How can machine learning and physics-informed neural networks complement traditional hydraulic and statistical modeling approaches?
- Why are data and instrument quality critical to effective computational pipeline monitoring?
- How can operators balance leak detection sensitivity, reliability, and time to detection?
- What makes complex gathering systems particularly well suited for new modeling approaches?
- How can better context help controllers distinguish a meaningful anomaly from normal pipeline behavior?
One theme runs throughout the discussion: effective leak detection requires understanding the specific problem you are trying to solve.
As Jerad explains in the episode, modern technology gives the pipeline industry new ways to model complex operating behavior—but no single approach solves every scenario. Strong leak detection programs increasingly rely on thoughtful combinations of technology, quality data, operational knowledge, and layered detection strategies.
For pipeline professionals evaluating new approaches to CPM, pipeline anomaly detection, machine-learning-based leak detection, or gathering-system monitoring, this is a conversation worth hearing.
Listen to Episode 451: “Leak Detection is Not Just One Problem with Jerad Stack” on the Pipeline Podcast Network.
