With the aim of improving the shortcomings of the traditional single hidden layer back propagation (BP) neural network structure and learning algorithm, this paper proposes a centrifugal pump performance prediction method based on the combination of the Levenberg–Marquardt (LM) training algorithm and double hidden layer BP neural network. MATLAB was used to establish a double hidden layer BP neural network prediction model to predict the head and efficiency of a centrifugal pump. The average relative error of the head between the experimental and prediction obtained by the double hidden layer BP neural network model was 4.35%, the average relative error of the model prediction efficiency and the experimental efficiency was 2.94%, and the convergence time was 1/42 of that of the single hidden layer. The double hidden layer BP neural network model effectively solves the problems of low learning efficiency and easy convergence into local minima—issues that were common in the traditional single hidden layer BP neural network training. Furthermore, the proposed model realizes hydraulic performance prediction during the design process of a centrifugal pump.

The two-dimensional incompressible Navier-Stokes equations and the SST k-ω turbulence model are used to investigate the aerodynamic performance of a wind turbine airfoil under clean and rough surface conditions.The DU 95-W-180 airfoil is analysized, which is widely used in wind turbines.The numerical simulation of the airfoil under clean surface condition is performed,and the numerical results have a good consistency with the experimental data.The studies in the effects of surface roughness on performance of the airfoil are mainly as follows:computation of the lift coefficient and the drag coefficient of the airfoil under different roughness heights on full surface and different roughness locations;the trends of the lift coefficient and the drag coefficient changing with the roughness heights and roughness locations;the critical value of roughness height and roughness location;the trends of the lift coefficient and the drag coefficient changing with the roughness heights at the critical locations.Furthermore, the unsteady flow field analysis of vortex shedding induced by the surface roughness is performed.

To study the effects of the performance of different types of impeller on the vortex pump, orthogonal test design, which is carried out by combining experimental test and numerical calculation, is adopted to optimize the type of design structure for the impeller in vortex pump. To find out the folding blade angle, the position of the folding point in the whole blade, and whether to wedge folding blade, an orthogonal test scheme with three factors and two levels is designed. A numerical simulation test is conducted for each scheme by analyzing the performance curve of orthogonal test plan to find the optimal performance of the program and analyzing the test data of each scheme to obtain the primary and secondary orders of the impact performance in the angle of folding blades of the vortex pump, the position of folding point of blades, and the wedge shape of blades. The results show that the optical blade type combination is the blade angle R30F60, the folding point is at 2/3 of the whole blade, and the blade does not adopt radial wedge. The optimal combination scheme is 36% higher than the design value at the rated flow head, the efficiency is 18.75% higher than the design value, the high-efficiency zone of the vortex pump is wider, and the performance meets the design requirements. Through orthogonal experimental design, the design cycle of vortex pump can be shortened effectively, the design level of vortex pump can be improved, and the hydraulic model with superior performance can be obtained.

No abstract