The paper describes the Robust Design Optimization (RDO) of an Electric Pump for
Dishwashers. The aim of this work is to apply RDO techniques to the design of a large scale
product and, in particular, the optimization target is to improve the efficiency and lower the
cost of the whole system, as well as to fulfil requirements on the overall dimensions and
head. Once an optimized geometry which meets all the requirements is obtained it will
become a standard pump listed in the company catalogue. The authors gave a particular
attention to the numerical model validations and to the theoretical study of the hydraulic
and electromechanical coupling. The parametric model representing the main geometrical
features of the water pump was created using the commercial CAD tool, Catia v5 and it had
some simplifications, compared with the model used for the production water pump but it
was considered to be accurate enough to provide a good analysis of the hydraulic
performance of the pump. A numerical model of the hydraulic part of the water pump was
created using the commercial mesh generator, ICEM CFD and CFX5 was used as
Computational Fluid Dynamics (CFD) solver. A Genetic Algorithm (GA) was used for the
Robust Design Optimization of the electric water pump and the best configuration was
selected using a Multi Criteria Decision Making (MCDM) tool over Pareto frontier designs.
In spite of the complexity of the operation necessary for the optimization of the electric
motor and hydraulic pump assembly, the optimized electric water pump is more efficient
than optimizing just the electric motor, and coupling it with an existing pump. The
optimized electric water pump will be validated by experimental tests in the near future.
