ONLINE IDENTIFICATION OF MODAL PARAMETERS USING ARTIFICIAL NEURAL NETWORKS

Using a neural network concept, art algorithm has been developed that can be employed to conduct on-line identification of modal parameters of a structure such as natural frequency, damping ratio, and mode shap e coefficients at measurement locations. Using a band-pass filter; the algorithm extracts from the measurement signal the frequency contents in the vicinity of a desired mode. The filtered signal is then used t o train a neural network which consists of a linear neuron with three weights. The structure of the neural network has been designed to allo w direct identification of modal parameters from the weights and to en hance efficiency for on-line implementation. The algorithm has been im plemented on a DSP (digital signal processor) system for performance e valuation. Modal parameter identification tests have been conducted fo r a laboratory circular plate structure. The algorithm runs approximat ely at a sampling frequency of up to 50 kHz for the DSP system used fo r the study and correctly identifies modal properties of the first two modes of the circular plate. This high performance on-line identifica tion algorithm is expected to be useful in improving the performance o f control systems that require knowledge of time-varying plant dynamic characteristics.