Processing Magazine

Multi-spectral IR flame detector delivers false alarm immunity

March 13, 2006

New FL4000 Intelligent Multi-Spectral IR (MSIR) Flame Detection System with neural network technology (NNT) gives process and plant engineers a potent new tool in protecting people, equipment and facilities from dangerous hydrocarbon flame sources. Until now, flame detection has been based on multi-spectral optical flame detectors integrated with expert signal processing. This type of system relies on an expert to design a fixed set of conditions. While these devices meet basic flame detection needs, they are prone to false alarms caused by reflected sunlight, arc welding, hot piping, machinery, and other false alarm sources. The new FL4000 combines an advanced optical MSIR sensor array with NNT processing. NNT is based on artificial neural networks (ANN), which are mathematical models of biological neurons in the human brain that correlate given signal patterns with target flame conditions. The optical IR sensor array and the neural network function together as an adaptive and intuitive decision-making mechanism with boundless optimization capability. The result is a breakthrough flame detector that provides the industry’s most reliable discrimination between actual flames and nuisance false alarm sources.