Optimization of twisted blade of pump as turbine based on high dimensional surrogate model

Abstract

In order to improve the operation efficiency of the twisted blade pump as turbine (PAT), a medium specific speed PAT was selected as the research object. The variables of the twisted blade plane blade profile were defined, the twisted blade was transformed into three plane blade profiles, and the blade profiles were parameterized by MATLAB 9.7 software. MATLAB 9.7, CFturbo 2020 and Fluent 19.2 were used to build the support vector machine-high dimensional model representation (SVM-HDMR) surrogate model function for efficiency optimization of PAT. Genetic algorithm was run on MATLAB 9.7 to optimize the surrogate model function, and the optimized blade profiles were fed back. The optimization results were verified by numerical simulation and experiment. The results show that the simulation efficiency of the PAT after optimization at the design operating point is 3.51% higher than the efficiency of the PAT before optimization, and the output power is increased by 5.3%. The test efficiency of the PAT after optimization at the design operating point is 3.4% higher than the efficiency of the PAT before optimization, and the output power is increased by 5.1%.

Multi-objective optimization design in a centrifugal pump volute based on an RBF neural network and genetic algorithm

“…It was first proposed by John Holland in 1975 and has been widely used in machine learning and optimization design research since its introduction in. 21 In multi-objective optimization design, using GA to find the optimal solution can avoid the complex mathematical solution process. It must produce only the corresponding objective function and fitness function and modify the probability of the optimization population through genetic operators selection, crossover, and mutation to obtain the optimal solution to the problem.…”

Multi-Objective Parameter Optimization of Submersible Well Pumps Based on RBF Neural Network and Particle Swarm Optimization

“…Tong Zheming et al [17] selected three optimization variables, namely impeller outlet width, diameter, and angle, and combined the Latin hypercube sampling method, BP neural network, and NSGA-III algorithm to optimize a model of a low specific speed centrifugal pump with pump head and efficiency as the optimization objectives. Jiang Bingxiao et al [18] optimized the geometric parameters of the vane profile of a centrifugal pump based on an intelligent algorithm, and the numerical simulation and experimental values of pump head efficiency were improved after optimization. Hao Zongrui et al [19] studied a water jet propulsion pump with lift resistance ratio and pressure as the optimization objectives and carried out optimization based on an improved particle swarm optimization algorithm, after which the lift resistance ratio was improved by 14.7 percentage points and the minimum pressure was increased by 20%.…”