Design of the flame detector filter vith wavelet transform and fuzzy logic algorithm [Dalgacık dönüşümü ve bulanık mantık çıkarım algoritması ile alev algılayıcı filtre dizaynı]

Official URL: https://dx.doi.org/10.1109/ELECO51834.2020.00008

The main problem of radiation-based flame detection systems is false alarms. The common reasons for false alarms of these systems are noise as interference on channel and environment, and other non-flame source's radiation that comes to the system. The aim of that study is to design a software-based filter that detects "fire" for incoming signal in the range of 5 Hz - 10 Hz and lasts at least 0.5 seconds. For processing the signal and decision-maker, Wavelet Analysis and Fuzzy Logic Algorithm has been used. The test was completed by using flame data on Simulink and applying the filter on Matlab & Simulink. The experiment results show that by using Wavelet Analysis, it is possible to extract signals within the specific frequency range from the original signal by decomposing it and after processing part, classifying the inputs for flame detection purposes.