Narrow Results

It is a crucial topic to identify the cross-correlations between time series in multivariate systems. In this paper, we extend the detrended cross-correlation analysis (DCCA) into the multivariate systems, assigned multivariate detrended cross-correlation analysis (MVDCCA). Numerical simulations of synthetic multivariate time series generated by two-exponent and mix-correlated ARFIMA processes are applied to illustrate the validity of the proposed MVDCCA. Results show that the external coupling parameter determines the strength of cross-correlation no matter that it is inter-independent or correlated among channels in a certain multivariate time series. The MVDCCA method is robust enough to detect the scale properties of time series by estimating the Hurst exponent. And we use cross-correlation coefficient to quantify the level of cross-correlations clearly. Furthermore, the MVDCCA method performs well when applied to the stock markets combining the stock daily price returns and trading volume of stock indices. By comparing results only using stock daily price returns in published literatures, we find that the higher recognizability between the pair stock indices can be observed whatever from the same regions or different regions in multivariate situations and the conclusions are more comprehensive.

To evaluate the safety of casing string is an important task in the oil exploitation. In this paper, the casing string with complex environment is investigated and the global sensitivity analysis (SA) technique is employed to identify the influential factors on the safety. Since the damage of casing string is of different kinds, three failure modes are mainly considered in the analysis. Then, the multivariate global SA technique is employed to identify the influential factors for the three failure modes simultaneously. Due to the full-size FE analysis of casing string which involves contact analysis of tread, being computationally expensive, a simplified model with full constraints are constructed. Then, to compute the multivariate global sensitivity efficiently, the neural network which is used to surrogate the FE model is employed to perform SA.