A suite of Python 2.7 scripts for automatically designing optimal radial turbine blades according to specific performance criteria (flow rate, efficiency) and under specified geometrical boundary conditions (inlet and outlet dimensions).

It relies on pyHT017 , a Python wrapper of the proprietary HT017 Fortran 77 code for quasi-3D flow simulation, and on an implementation of the Particle Swarm Optimization algorithm with function evaluations spread across multiple cores using the Python multiprocessing module, in order to explore the design space and automatically converge towards the globally optimal design variant.

This set of pieces of software was originally written as part of a project with the goal of automatically designing multiple variants of a turbine with maximized efficiency at different target flow rates, and continued to evolve over a period of a few years, finding application as a “workhorse” not only in the design of new variants of existing turbines, but also in the research of the limits of turbine design for drastic new requirements.

Its application was made public with the paper “Design of Radial Turbine Meridional Profiles using Particle Swarm Optimization” at the 2^nd International Conference on Engineering Optimization (EngOpt2010).