Methodology of flow rate assessment of submerged hydraulic ballast pumps on modern product and chemical tankers with use of neural network methods
Andrzej Banaszek
Abstract
The paper presents a methodology of flow rate assessment of submerged ballast pumps with hydraulic drive mounted on board modern product and chemical tankers. The control of flow rate of hydraulic ballast pumps are different from other popular solutions because based on the the use of constant torque controller STC type on the pump hydraulic drive motors referring to standard solutions with pump speed controllers. The method is based on the use of Artificial Neural Network system algorithms with Fitting app of Matlab R2020b computer program. The formulated calculation problem concerns the determination of the corrected flow rate of ballast pumps using only the base flow and drive characteristics of the analyzed pump SB300 and the adjustment value of hydraulic constant torque controller was presented. Shows the structure of the algorithm and the results of numerical experiments for ballast pump characteristics with different adjusted oil pressure drop values in the hydraulic pump motor. Advantages of using neural network algorithms are presented.