Research and Development
Below are three wave forms of flames, as recorded on the field. These graphics illustrate the flicker frequencies detected by our system, in the 900 to 1700 nanometres infrared region.
Figure 1 – IR signal of a pulverized coal flame (Intensity against time) |
|
 |
|
Figure 2 – IR signal of a natural gas flame |
|
 |
|
Figure 3 – IR signal of a heavy oil flame |
|
 |
Most of time, these high frequencies are considered as noise, and not as a part of the flame signal. For its part, our team has discovered that the high IR flicker frequencies are very present and stable, and are an integral part of the flame.
Until now, the flame detection technology used in the combustion industry did not permit to verify if there were high frequencies in flames. Also, we developed flame detectors able to read these frequencies, which ensure a better characterization of the flame, a more precise discrimination, and a better adjustment.
PIA still pursues researches about high flicker frequencies in infrared signals and, based on the results obtained with our analysis, we are currently developing digital filters, resulting from a Fourier-like decomposition adapted to the random signals of a flame (see figure no. 4). The digital filters, user controlled, will permit to associate a flame to a precise IR frequency range. This technology will be particularly useful in multi-fuel combustion.
Figure 4 – Example of digital filter |
|
 |
Spectral curves of sensors and fires produced by different fuels
 |
|
Spectral curves of the UV tube and IR cell in scanner head Window 141™.
|
|
 |
|
Spectral curves of 3 different burning fuels.
|