Intelligent Alarm Classification Using Paraconsistent Logic Applied to Pressure Monitoring

Abstract

To ensure the safety of operations in storage tanks, it is essential to have accurate alarms generated by sensors. Often, contradictory or incomplete information causes false alarms. This study presents a new way to classify alarms using Paraconsistent Annotated Logic with Annotation of Two-Value (PAL2v). The method defines positive evidence levels for hazardous and normal conditions based on pressure readings, which are analyzed by a Paraconsistent Analysis Node (PAN) that classifies them into “Act immediately” (for clear, critical cases), “Report instrument problem” (for sensor faults or data errors), and “Alarm with a possibly incorrect setpoint” (for ambiguous alarm patterns or low-certainty signals).