Development of optimally disordered critical random excitation

The problem of determining the critical power spectral density (PSD) functi on of a partially specified stationary Gaussian load process which maximize s the response of a linear system has been considered. The partial specific ation of the load is given only in terms of its total average energy. The c ritical input PSD turns out to be highly narrow banded which fails to captu re the erratic nature of the excitation. Consequently, the trade-off curve between the maximum linear system response and the disorder in the input pr ocess, quantified in terms of its entropy rate, has been generated. The Par eto optimization theory is used to tackle the conflicting objectives of sim ultaneous maximization of the system response and the input entropy rate. C onsequently, the non-linear multi-objective optimization has been carried o ut using a Multi-criteria Genetic Algorithm scheme. An illustrative example of determining the critical input of an axially vibrating rod excited by a partially specified stationary Gaussian load process has been considered. (C) 2001 Academic Press.