Aircraft Cargo Compartment Multisensor Smoke Detection Algorithm Development

摘要

There is a need to effectively develop and test an advanced fire detection system for aircraft cargo compartments that significantly reduces false alarms and improves alarm time response. Title 14 Code of Federal Regulations Part 25.858 requires that aircraft detection systems alarm within 1 minute of the start of a fire. Gas concentrations, temperature fluctuations, and particulate levels are three main parameters representative of a complete fire signature. Current aircraft detection systems depend solely on one parameter, particulate levels, for the detection of this wide fire signature. Improved fire detection capabilities can be achieved by combining multiple fire signatures or parameters in specific algorithms. An advanced fire detection system combining an ionization smoke detector, thermocouple, smokemeter, and a carbon monoxide (CO)/carbon dioxide (CO2) gas probe was installed in a Boeing 707 forward cargo compartment. A broad spectrum of fire and nuisance sources were tested to produce a matrix of extreme detector levels from all four sensors. This matrix provided alarm threshold criteria that aided in the development of a multisensor algorithm based on fire signatures such as CO and CO2 gas concentrations, temperature, ionization chamber voltage, and percent light transmission per foot. Multiple algorithms were created to determine the most effective multisensor algorithm that responded the fastest to fires while providing nuisance immunity.