Title: Improving Modal Parameter Estimation by Complementary Output–Output Relations
Abstract: Frequency domain modal parameter estimation from input–output data requires direct measurement or estimation of the input and output signals. In different applications those measurements, especially the excitation signals, are difficult to obtain and/or the assumptions could be poor or inappropriate (high uncertainty or high levels of noise). In this situations, the output–output relations can be used as auxiliary or complementary equations. The current work presents a framework for the identification of modal parameters estimation using maximum likelihood estimation incorporating the output–output relations in addition to the input–output ones. Since the output–output relations are independent of the input signals and its related uncertainty they will improve the system estimation. The ML estimator presents properties of consistency and efficiency and converge to the noiseless solution, but it involve calculating the inverse of the covariance matrix. An extended practice is to consider or assume that the noise of the frequency responses is uncorrelated, in this sense the covariance matrix becomes diagonal and the computational time is reduced. However, the price to pay is that the efficiency of the estimator is altered (the estimator does not reach the Cramer-Rao lower bound). The current work shows that incorporating the output–output relations to the input–output set of equations generates results closer to the ML estimator with a reduced computational load.
Publication Year: 2017
Publication Date: 2017-01-01
Language: en
Type: book-chapter
Indexed In: ['crossref']
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot