Narrow Results

A so-called FuzzBEM methodology for analyzing the influence of uncertain acoustic and structural parameters on the radiated sound field of vibrating structures combining fuzzy arithmetic and fast multipole boundary element method is introduced. Uncertainties in acoustic properties may result from uncertain parameters of the vibrating mechanical structures, e.g. material density or geometry, as well as from uncertainties in the acoustic domain, e.g. sound velocity. The use of the transformation method in the proposed approach allows to employ simulation tools based on the crisp number arithmetic by appropriate preprocessing of the fuzzy numbers modeling the uncertain input parameters and postprocessing of the simulation results to determine the fuzzy numbers for the considered output quantities.

In this contribution, the proposed FuzzBEM procedure is applied to a sound radiating, vibrating stiffened cylindrical shell where the investigated uncertainties include the shell wall thickness and the driving frequency of a monofrequency point load and the air density and sound velocity. As exemplary output quantities of acoustic performance, the acoustic pressure at multiple field points and the radiated sound power are evaluated.

The proposed coupling of fuzzy arithmetic and acoustic boundary elements yields run times two orders of magnitudes or more longer than a single BEM calculation. Nevertheless, the systematic parameterization obtained by the proposed fuzzy analysis has the potential to reveal input–output relationships difficult to identify with individual conventional BEM simulation runs.

Partial Discharge (PD) will lead to the decomposition of SF₆, and its decomposition characteristics under different kinds of PDs are significantly different. Thus, it is of great significance to diagnose and evaluate the GIS operating state by analyzing the SF₆ decomposition products to detect PD. In this paper, the long-term sustainable discharge decomposition experiment is conducted on needle-plate defect model in order to simulate the whole process of GIS insulation deterioration. SO₂F₂, SOF₂, SO₂, CO₂ and CF₄ are selected as feature components. Concentration of SO₂, rate of (SO₂F₂+SOF₂+SO₂) and the ratio of (SO₂F₂+SOF₂+SO₂)/(CO₂+CF₄) are proposed as feature parameters and their physical significances are also explained. Combined with the variation of three characteristic parameters with PD development, fuzzy analysis is used to identify the severity of discharge. Then corresponding fuzzy judgment model is established and divides GIS insulation deterioration trend into three states of “normal stage”, “deterioration stage” and “saturation stage”, and it performs well in this classification. The corresponding theoretical basis is provided for GIS insulation fault diagnosis and oncondition maintenance.