Multi-Objective Optimization of Jet Pump Based on RBF Neural Network Model

Abstract

In this study, an annular jet pump optimization method is proposed based on an RBF(Radial Basis Function) neural network model and NSGA-II(Non-Dominated Sorting Genetic Algorithm) optimization algorithm to improve the hydraulic performance of the annular jet pump applied in submarine trenching and dredging. Suction angle, diffusion angle, area ratio and flow ratio were selected as design variables. The computational fluid dynamics (CFD) model was used for numerical simulation to obtain the corresponding performance, and an accurate RBF neural network approximate model was established. Finally, the NSGA-II algorithm was selected to carry out multi-objective optimization and obtain the optimal design variable combination. The results show that the determination coefficient R2 of the two objective functions (jet pump efficiency and head ratio) of the approximate model of the RBF neural network were greater than 0.97. Compared with the original model, the optimized model's suction angle increased, and the diffusion angle, flow ratio and area ratio decreased. In terms of performance, the head ratio increased by 30.46% after the optimization of the jet pump, and efficiency increased slightly. The proposed jet pump performance optimization method provides a reference for improving the performance of other pumps.

Побудова Номограм Для Визначення Характеристик Струминного Насоса При Його Обертанні В Свердловині

“…Графічно даний розрахунковий етап ілюструється пошуком точки перетину згаданих характеристик і визначає координати робочої точки свердловинної насосної установки: відносний напір та відносну витрату. Поєднання даних режимних параметрів в подальшому застосовується для розрахунку енергетичної характеристики струминного насоса [9]. У випадку застосування одно-або двоконтурних ежекційних систем додатково розраховуються значення витрат в замкнутих контурах циркуляції промивального розчину [10].…”

Path Planning Optimization for Driverless Vehicle in Parallel Parking Integrating Radial Basis Function Neural Network

“…Through analysis, the Bézier curve needs at least six control points to satisfy curvature continuity constraint, that is, at least fifth order. Because the higher the degree of the Bézier curve, the more complex the curve [17], the fifth order Bézier curve shown in Equation ( 19) is selected to fit the initial planning path of arc-line-arc. Bézier curve is a smooth and continuous engineering curve, which can describe complex shape and show curve trend through control points.…”

Multi‐objective optimization of annual electricity consumption and annual electricity production of a residential building using photovoltaic shadings

“…For generating these data, random sampling is applicable which will subsequently use to assess objectives. There are multiple random sampling methods used in various literatures, and one of the best is the Latin Hypercube Sampling method (LHS) which is applied by many researchers in various fields 63‐67 . An ideal choice should be finally take into account from random data.…”

“…There are multiple random sampling methods used in various literatures, and one of the best is the Latin Hypercube Sampling method (LHS) which is applied by many researchers in various fields. [63][64][65][66][67] An ideal choice should be finally take into account from random data. There is usually too much difficulty in finding the ideal choice, and as a result, it would be better to use the help of machine learning and optimization algorithms.…”