Neural network and multi-objective optimization of confined flow characteristics on circular cylinder in standing double vortex region

Senthilkumar, R.; Vasudevan, Premalatha; Sethuramalingam, Prabhu

doi:10.1007/s00521-020-05079-z

Cited by 5 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A neural network 30,31 is a mathematical technique based on the natural neural network in the human brain. To interpolate a function, a network with three levels (input, hidden, and output) is built and the connections between them are weighted, as shown in Figure 3.…”

Section: Neural Network Modelmentioning

confidence: 99%

Research on cooperative optimization of multiphase pump impeller and diffuser based on adaptive refined response surface method

Peng

Zhang

Zhiguang

et al. 2022

Advances in Mechanical Engineering

View full text Add to dashboard Cite

Multiphase pumps play an important role in the exploitation of natural gas hydrate. Compared with ordinary pumps, they can handle fluids with higher gas volume fraction (GVF). Therefore, it is important to improve the performance of the pump under high GVF. A model pump is designed based on the design theory of axial flow pump and centrifugal pump inducer. The hydraulic performance of the model pump is verified by numerical simulation and experiment. The Sparse Grid method is applied to the design of experiment (DOE), and three different adaptive refined response surface methods (RSM) are applied to the build the approximate model. Refinement points and verification points are used to improve and verify the precision of the response surface, respectively. The model with high precision and high computational efficiency is obtained through comparison and analysis. The multi-objective optimization of the optimal response surface model is carried out by MOGA (Multi-Objective Genetic Algorithm) method. The pressure increment of the optimized model is increased by 38 kPa. The efficiency is significantly improved under large mass flow conditions. The hydraulic performance of the optimized model is compared with that of the basic model. And the reasons that affect the performance of the multiphase pump are analyzed.

show abstract

Section: Neural Network Modelmentioning

confidence: 99%