Fast Design Procedure for Turboexpanders in Pressure Energy Recovery Applications

Abstract

Sustainable development can no longer neglect the growth of those technologies that look at the recovery of any energy waste in industrial processes. For example, in almost every industrial plant it happens that pressure energy is wasted in throttling devices for pressure and flow control needs. Clearly, the recovery of this wasted energy can be considered as an opportunity to reach not only a higher plant energy efficiency, but also the reduction of the plant Operating Expenditures (OpEx). In recent years, it is getting common to replace throttling valves with turbine-based systems (tuboexpander) thus getting both the pressure control and the energy recovery, for instance, producing electricity. However, the wide range of possible operating conditions, technical requirements and design constrains determine highly customized constructions of these turboexpanders. Furthermore, manufacturers are interested in tools enabling them to rapidly get the design of their products. For these reasons, in this work we propose an optimization design procedure, which is able to rapidly come to the design of the turboexpander taking into account all the fluid dynamic and technical requirements, considering the already obtained achievements of the scientific community in terms of theory, experiments and numeric. In order to validate the proposed methodology, the case of a single stage axial impulse turbine is considered. However, the methodology extension to other turbomachines is straightforward. Specifically, the design requirements were expressed in terms of maximum allowable expansion ratio and flow coefficient, while achieving at least a minimum assigned value of the turbine loading factor. Actually, it is an iterative procedure, carried out up to convergence, made of the following steps: (i) the different loss coefficients in the turbine are set-up in order to estimate its main geometric parameters by means of a one dimensional (1D) study; (ii) the 2D blade profiles are designed by means of an optimization algorithm based on a “viscous/inviscid interaction” technique; (iii) 3D Computational Fluid Dynamic (CFD) simulations are then carried out and the loss coefficients are computed and updated. Regarding the CFD simulations, a preliminary model assessment has been performed against a reference case, chosen in the literature. The above-mentioned procedure is implemented in such a way to speed up the convergence, coupling analytical integral models of the 1D/2D approach with accurate local solutions of the finite-volume 3D approach. The method is shown to be able to achieve consistent results, allowing the determination of a turbine design respectful of the requirements more than doubling the minimum required loading factor.