Understand anomaly detection

The Batch Tracking application allows anomalies to be tracked in the pipeline. The movement of the anomalies is simulated based on many factors in the pipeline, such as the flow of the pipeline segment and the anomaly’s slip factor. Anomalies can be logical such as a marker for an interface between two batches, or physical such as a pipeline scraper (also known as a pig).

An Anomaly Detect (adetect) record can be configured via the Batch Tracking Compiler (BTC) at nodes within the pipeline equipped with anomaly detection capabilities. Each Anomaly Detect record should specify one and only one anomaly type it can detect.

When an anomaly passes through a detector in the pipeline, a detection signal should be communicated to the Anomaly Detect record so that it can sync up the simulated location of the anomaly with the actual location of that anomaly in the pipeline. This can be done in one of the following ways:

1. An external application updates the timestamp array of the Anomaly Detect record to capture a detection event.

2. The trigger point associated with an Anomaly Detect record is set to a target state (via a pulse signal), which then causes an update to the timestamp array of the Anomaly Detect record to capture a detection event. The detection trigger can be a status point, a multistate point or an Advanced Logic Processor (ALP) record.