ABSTRACT

Slurry density is an important parameter that indicates the performance of dredging operators. Currently, nuclear-based gamma densitometer is the most widely used instrument for measuring slurry density in cutter suction dredgers. The use of nuclear sources may cause problems of operation, safety, and cost. Therefore, this study aims at proposing a computational model for calculating the slurry density in a non-nuclear manner. To this end, real dredging data were collected to train a supervised back-propagation neural network, which can be considered as the calibration stage of a densitometer. In order to improve the computational accuracy of the neural network, we present a revised Gaussian filter to process the original input data for the neural network. From the measurement results, it can be concluded that the proposed model is effective and feasible.