Continuous floc image analyser (C-FIA) for tracking floc particle dynamics during coagulation–flocculation–settling processes

Abstract

A continuous floc image analyser (C-FIA) was developed for tracking the dynamics of floc particle formation, growth and settling during conventional drinking water treatment processes, comprising coagulation, flocculation and settling. For this, a script was developed to (1) extract the video frames and sync these with reaction time (t), (2) transform the video frames to an equivalent matrix of pixel RGB data, (3) optically filter the video frames for background effects, (4) calculate the filtered (corrected for background) flocculation index (C-FIA FI_t) for the frame belonging to each process time through RGB data and (5) correct the C-FIA FI_t for baseline RGB signal (raw water frame, C-FIA FI₀). The video frames collected by C-FIA have potential to provide direct information on floc size and concentration. The C-FIA FI data were acquired from jar tests using a range of alum doses performed on three natural surface water samples with highly contrasting water quality. The C-FIA FI data acquired, displayed lower variance in comparison with available data of photometric dispersive analysis (PDA). The indices previously developed for PDA analysis were utilized in this study for C-FIA FI data, including plateau index. In experiments conducted, this index showed an inverse relationship with residual turbidity, UVA₂₅₄ and colour following coagulation. This indicates a potential for C-FIA to be applied for coagulant dose control at conventional drinking water treatment plants. On the other hand, from the experimental work performed, impacts of light intensity and light radiation angle to the photography cell on the resulting FI value were found. These can impact on the pixels' red/green/blue (RGB) values of the video frames/images, and consequently can impact on the resultant C-FIA FI data.