Sludge dewatering performance evaluation method and system based on multi-modal data fusion

Abstract

The application discloses a sludge dewatering performance evaluation method and system based on multi-modal data fusion, and relates to the technical field of sludge dewatering monitoring and intelligent evaluation; the method comprises the following steps: synchronously collecting the surface image of the sludge cake of a belt filter press in an evaluation window, the transverse correction position of the filter belt, the spraying cleaning pressure, the filtrate state and the running speed of the filter belt, and determining the filter belt cycle phase; based on the surface image of the sludge cake, a transverse water lag zone is extracted, the transverse projection, the transverse center position of the water lag zone, the transverse expansion scale of the water lag zone and the water lag area ratio are obtained; the memory drift index is calculated in combination with the corresponding information of the target historical moment; the spraying deficiency and the filtrate deterioration are further extracted, and the disturbance amount of the drift zone is calculated; and the dewatering non-uniformity is calculated based on the transverse projection, and the sludge dewatering performance evaluation value and the drift zone dominance are determined. The application can identify persistent transverse water lag abnormalities, and improve the accuracy of sludge dewatering performance evaluation and the pertinence of operation and maintenance analysis.

Images