Anti-vibration optimization of the key components in a turbo-generator based on heterogeneous axiomatic design

With the increasing demand for the cleaner production of electric energy, many countries around the world strive to develop new turbo-generators with advanced parameters and large capacities. In the design of a new turbo-generator, it is important to optimize the anti-vibration performance of its key components in order to ensure its safe and reliable operation. The anti-vibration optimization of "the key components in a turbo -generator when a set of alternative schemes are provided for selection is a heterogeneous multi-attribute decision making (HMADM) problem involving deviation attributes, which cannot be solved by present HMADM approaches. To solve such a HMADM problem, a novel heterogeneous axiomatic design (HAD) method is developed by introducing the distance measure into axiomatic design for computing the information contents of the alternative schemes. With the attribute data in various mathematical forms unified as triangular fuzzy numbers, their probabilities of success can be conveniently calculated by a unified formula once their corresponding positive and negative ideal solutions are determined. As a result, the complexity and computational cost involved in computing the information contents of alternatives can be greatly reduced, which enables the efficient solution of the HMADM problem involving deviation attributes. A case study on the fixture scheme optimization for the stator end windings in a large turbo generator demonstrated the feasibility and effectiveness of proposed HAD. The ninth fixture scheme with the structural manufacturability being very good (VG), the natural frequencies at the turbine and exciter ends being [68.2, 78.5]Hz and [72.5, 83.2]Hz, the maximum amplitudes at the turbine and exciter ends being 67 pm and 72 pm was chosen as the optimal scheme from nine alternatives according to the HMADM results. (C) 2016 Elsevier Ltd. All rights reserved.