ADVANCED INFRARED SYSTEMS FOR DETECTION OF BUILDING FIRES

FIRES KILL NEARLY 10,000 PEOPLE EACH YEAR IN THE U.S. AND CAUSE NEARLY 300,000 INJURIES AND OVER 6 BILLION DOLLARS IN PROPERTY DAMAGE. THE PER CAPITA LOSS OF LIFE DUE TO FIRES IN THE U.S. IS HIGHER THAN FOR ANY OTHER INDUSTRIAL COUNTRY. CONSEQUENTLY, THERE IS A GREAT NEED FOR ADDITIONAL RESEARCH ON FIRES AND FIRE SAFETY. IT HAS LONG BEEN RECOGNIZED THAT MULTIPARAMETER DETECTORS WHICH CAN MEASURE, FOR EXAMPLE, HEAT, SMOKE AND CO ARE MORE RELIABLE INDICATORS OF A REAL FIRE. THERE ARE THREE RECENT EVENTS WHICH TOGETHER PROVIDE AN OPPORTUNITY TO DEVELOP A NEW GENERATION OF FIRE DETECTIN SYSTEMS, 1) THE AVAILABILITY OF LOW COST, HIGH SPEED COMPUTATIO IN THE FORM OF PERSONAL COMPUTERS; 2) THE AVAILABILITY OF SOFTWARE PACKAGES INVOLVING TECHNIQUES SUCH AS NEURAL NETWORKS OR FUZZY LOGIC; 3) THE AVAILABILITY OF ADVANCED SENSOR TECHNOLOGY WHICH CAN DRAMATICALLY INCREASE, AT RELATIVELY LOW COST, THE AMOUNT OF INFORMATION UPON WHICH AN ALARM CONDITION CAN BE ESTABLISHED. THE OBJECTIVE OF THE PHASE 1 PROPOSAL WOULD BE TO INVESTIGATE THE FEASIBILITY OF USING OPEN-PATH OR EXTRACTIVE FT-IR MEASUREMENTS FOR AN ADVANCED FIRE DETECTION SYSTEM. THE PHASE 1 PROGRAM WOULD HAVE THE FOLLOWING SPECIFIC OBJECTIVES; 1) MODIFY AN FT-IR SPECTROMETER SYSTEM TO ALLOW OPERATION IN THREE DIFFERENT DETECTION MODES (SHORT PATH, LONG PATH, AND EXTRACTIVE); 2) MAKE MEASUREMENTS ON A STANDARD FLAME IN AN OPEN ROOM TO EVALUATE THE DETECTION OPTIONS; 3) USE ADVANCED SIGNAL PROCESSING TECHNIQUES TO DEFINE WHEN A FIRE CONDITION IS PRESENT; 4) IDENTIFY A PROTOTYPE, FT-IR BASED FIRE DETECTION SYSTEM WHICH WOULD BE BUILT AND TESTED UNDER THE PHASE 2 PROGRAM.