Model of information and measuring control system of centrifugal cargo pump on a chemical tanker of the Ivan Poddubny type

Abstract

On chemical tankers, the main cargo devices are pumps, so it is necessary to constantly monitor their operating conditions and performance indicators during operation. The article discusses modeling using MATLAB/Simulink software of a single-stage horizontal centrifugal pump, which can be used to create an information-measuring system for monitoring the performance of cargo pumps using the example of a chemical tanker of the Ivan Poddubny type. An analysis of the modeling results is carried out, including an assessment of pump efficiency, pressure loss and other indicators in various operating modes. The results allow us to draw conclusions about the operation of the pump under various loads and variable conditions. The simulated pump worked with seven different liquids pumped in chemical tankers: ethyl alcohol, N-propyl alcohol, phenol, chloroform, castor oil, 55% nitric acid and water. After this, a neural network was created using the MATLAB/Neural Network Fitting Tool application. The input data of the network were volumetric flow, pressure, shaft power, torque and suction pressure. The output value was the pump efficiency, the value of which was estimated for each fluid. The root mean square error of approximation was very close to zero. According to these results, the artificial neural network showed satisfactory results for predicting the efficiency of the centrifugal cargo pump of the Ivan Poddubny type chemical tanker; therefore, it can be used in ship information and measurement systems. Overall, this paper represents a valuable research contribution to the field of modeling horizontal centrifugal pumps using Simulink that can be useful to engineers and researchers involved in the design and optimization of these pumps.