Research On Automatic Design Optimization Of Mixed Flow Impeller

The impeller is the core component of the mixed-flow pump; therefore, its performance has direct impact on the operational efficiency, stability and cavitation performance of the pump. Based on computer and computational fluid dynamics (CFD) technology, which have been widely used in computer aided design (CAD) and performance prediction of mixed-flow pump, the automatic optimization design research of mixed-flow pump impeller, which integrates CAD and CFD technology effectively, can truly realize optimization design of mixed-flow pump impeller, further improve mixed-flow pump performance, reduce the design cost, and shorten the development cycle.This dissertation, based on the multidisciplinary design optimization software iSIGHT-FD, constructs a platform for automizing the optimum design of mixed-flow pump impellers by combining CAD of mixed-flow pump impeller and CFD simulation technique with the numerical optimum technology. The platform runs the automation of the optimum design by using mathematical models for controlling of the blade design direction and taking the existing mixed-flow pump impeller model as the object. The CFX software is used to simulate the internal three-dimensional flow field of the mixed-flow pump impeller, and calculate the characteristic parameters of the model impeller. By analyzing the methods of CFX input and output data, the format of the pre-processing input data and post-processing output data are determined; four parameters of the optimization including the chord/pitch ratios of the blade tip and the blade hub, as well as the spanwise dimensionless load coefficient A/B and C/B are chosen for controlling the length of each section of the airfoil and the vane setting angle of the airfoil; the parameterized modeling of the impeller blade is realized through adjusting the CAD program of the mixed-flow pump impeller. By analyzing optimization parameter limits of the impeller blade modeling and grid partition, the effectiveness of the optimization parameter search space is verified. Finally, the automatic optimum design process of mixed-flow impeller is constructed including the parameterization modeling, grid partition, numerical optimization and the optimum methods of NLPQL through the software iSIGHT-FD, to carry out the automatic optimum design for the initial blade of the design duty.By comparing the performances of the impeller blade with and without optimization, it is obvious that the performance of the optimized impeller is improved, which shows that this automatic optimum design method is effective and feasible.