Geochemical anomalies are the basis of mineral deposit prediction. Through the study of the characteristics of geochemical anomalies, we found that their distribution was consistent with a generalized Pareto distribution (GPD). In the present study, we designed a model for geochemical anomaly extraction via a GPD. In the designed GPD model, we used the kurtosis method to estimate the threshold value of the GPD. Furthermore, a principal component analysis (PCA) was used to extract comprehensive information of different geochemical elements in which minerals are enriched. On this basis, a new algorithm named the GPDA model was designed for deep mineral prediction by using the GPD and PCA, and the methods of the GPDA for selecting parameters were studied. The study data for Ba, Pb, As, Cu, Au, Mo, Co, and Zn originated from 26 exploration lines of the Jiguanzui Cu-Au mining area in Hubei, China. The proposed GPDA model was applied to deep mineral prediction in the study area. We estimated the parameters of the GPDA model, and the thresholds of Ba, Pb, As, Cu, Au, Mo, Co, and Zn were 457.8612, 56.1823, 28.8454, 910.1272, 89.4283, 34.5267, 84.9445, and 121.4863, respectively. The comprehensive information threshold value was 0.4551. The comprehensive abnormal distribution area of geochemical element contents was obtained from thresholds. The results showed that the method used to identify abnormal areas was consistent with the range of ore bodies identified by actual engineering exploration, demonstrating that the GPDA model was effective. Finally, we predicted that there was a new blind ore body located at a depth of about 1100 m below ground between drill holes KZK10 and KZK11. The results have important theoretical and practical significance for deep ore prospecting.

A novel heterogeneous behavior representation for linear stochastic switched system is proposed in the discrete-time domain. The switching modes of state evolution and measurement output are described by two random sequences with known probability information. The more general and flexible framework covers several classes of well-studied models as special cases, and can be served to manage different complex systems with random abrupt changes in structure and parameter, so that it has wider applicability than existing models. By introducing an equivalent auxiliary system in virtue of the mode-dependent random parameter matrices, the filter design schemes, including an optimal and a suboptimal recursive algorithms, are performed for the established model in the minimum mean square error sense to meet different application requirements. Illustrative numerical examples demonstrate the effectiveness of the proposed formulation and the corresponding filters that enjoy a promising application prospect.INDEX TERMS State estimation, minimum mean square error, heterogenous, switching modes, random parameters matrices.