Applying Neural Network Technology To Next Generation Flame Detectors

Accurate and reliable industrial flame monitoring has always represented a technical challenge. The main flame monitoring concern lies in the requirement to differentiate spectral radiation emitted by flames from that of background radiation, which is always present in the industrial environment. Insufficient differentiation of flames from background sources typically causes the highly undesirable condition of false alarm. When a false alarm condition occurs, it activates automated fire suppression equipment, causes operational interrupts, consumes extinguishing materials, and requires resolution of the alarm condition, which can include a complete system shutdown and restart.

Until recently, the state-of-the-art in optical flame detection was based on multi-spectral detection integrated with an expert signal processing system. This expert system is typically based on a fixed set of conditions derived from the spectral analysis of a selected group of flames and programmed into computer logic. With proper installation, these types of instruments fulfill basic flame