In big thermal power plants, mixed flow pumps are commonly utilized for cooling water operations. In mixed flow pump impeller design, meridional streamlines are estimated by dividing the annulus by equal area. Inlet and output blade angles are determined empirically based on speed. Before estimating meridional streamlines, similar coefficients are utilised to decide impeller architecture. With inlet and exit angles set, blade pieces are placed out on stream surfaces. Adjusting the blade portions on the stream surfaces controls the composite blade shape. The above industrial method uses empirical relationships and design constants. The industrial design method ignores what happens in the pump flow tube, making it a poor guide for new pump design and development. The designer has less influence over events in the aforementioned design process. Lack of rational basis hinders manufacturing deficiencies in expected performance.
This book is written to solve the foregoing obstacles, one must create a reasonable basis for impeller design starting from basic principles, minimizing empirical co-relations. Such a design from basic principles gives the creator more control over the outcome of his design while keeping the physical concept in mind and allowing him to correct any pump performance issues. Considering the foregoing, a mixed flow pump impeller is designed utilizing fluid mechanics and turbo-machinery relations. After the impeller design is complete, modelling software is used to draw its 3-Dimensional model, ANSYS analysis software is used to estimate the stress under uniformly distributed and uniformly varying load, and natural frequency is calculated. Simulation of the same is performed to determine the design’s acceptability. Last but not least, the artificial neural network method utilized to assess impeller blade results is a superior choice than normal trapezoidal blades.
