Answers are grounded in this paper and cite their sources.
Solving pipeline systems is very important for the design engineer. The traditional method for solving these systems is complicated and time consuming. A network is trained and verified to predict the flowing discharge from the friction loss per unit length, the pipe diameter, and ratio between the equivalent roughness height and pipe diameter. About 30,000 synthetic realizations are used to train, verify, and test this network. These realizations cover a wide range of pipe diameter, equivalent roughness height, and water velocity. Testing this network shows a very good agreement between the predicted and actual discharges. Matlab codes are created to use this network in solving the discharge of the three tanks problem, three tanks with pump, and four tanks with pump. The root mean squared error between this approach and Engineering Equation Solver (EES) software is about 6*10-4 m3/s. The results ensure the ability of this approach in solving pipeline systems easily, accurately and quickly.
No takes yet. Share an insight, caveat, or question.
Ahmed Tawfik (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: