Joint state and parameter estimation in particle filtering and stochastic optimization

In this paper, an adaptive estimation algorithm is proposed for non-linear dynamic systems with unknown static parameters based on combination of particle filtering and Simultaneous Perturbation Stochastic Approxi- mation （SPSA） technique. The estimations of parameters are obtained by maximum-likelihood estimation and sampling within particle filtering framework, and the SPSA is used for stochastic optimization and to approximate the gradient of the cost function. The proposed algorithm achieves combined estimation of dynamic state and static parameters of nonlinear systems. Simulation result demonstrates the feasibilitv and efficiency of the proposed algorithm

Source apportionment of polycyclic aromatic hydrocarbons(PAHs) in a sediment core from Lake Dagze Co, Tibetan Plateau, China: Comparison of three receptor models

Receptor models are a useful tool for identifying sources of polycyclic aromatic hydrocarbons(PAHs) in multiple environmental media. In this study, three different receptor models(including the principal component analysis-multiple linear regression(PCA-MLR), positive matrix factorization(PMF), and Unmix models) were used to apportion the sources of 16 priority PAHs in a sediment core of Lake Dagze Co. The ΣPAHs(sum of all 16 measured PAHs) concentrations ranged from 51.89 to 132.82 ng/g with an average of 80.39 ng/g. TheΣ PAHs were dominated by 2-3 ring PAHs, accounting for 80.12% on average, thereby indicating that they mainly originated from biomass and coal combustion and/or from long-range atmospheric transportation. The three models produced consistent source apportionment results. The greatest contributor to ΣPAHs was biomass combustion, followed by coal combustion, vehicle emissions, and petrogenic sources. Moreover, the temporal variation of the common sources was well-correlated among models. The multi-method comparison and evaluation results showed that all three models were useful tools for source apportionment of PAHs, with the PMF model providing better results than the PCA-MLR and Unmix models. The temporal trends of factor contributions were verified by PAHs with different ring numbers. Significant correlations were found between the simulated concentrations of each source factor and the PAHs with different ring numbers(P<0.01), except for the petrogenic source identified by the Unmix model(P>0.05). This study can provide useful information for further investigation of source apportionment of PAHs in the sediment cores.