Author(s): Haidong Yang; Dongguo Shao; Yi Xiao; Zhuomin Wang
Linked Author(s):
Keywords: Ccidental water pollution; Traceability; Bayesian inference; Differential evolution; Monte Carlo
Abstract: The inverse problem of source traceability of accidental water pollution events is considered as the Bayesian estimation problem. Firstly the posterior probability distribution of the source's position, intensity and events initial time are deduced with Bayesian inference. Then the marginal posterior probability density is obtained by sampling the posterior probability distribution using the differential evolution algorithm and Markov Chain Monte Carlo simulation. Further these unknown parameters of accidental water pollution events’ source are estimated so as to identify source term accurately. Finally this proposed algorithm is compared with Bayesian-MCMC algorithm by numerical experiments. The conclusions are as following: the differential evolution Markov chain Monte Carlo algorithm can reduce the iterations, improve the recognition accuracy, and reduce the overall average error obviously and it is more stable and robust than Bayesian-MCMC algorithm and can identify accidental water pollution events’ source effectively. Therefore, it provides a new idea and algorithm to solve the difficulty of traceability tracing problems in accidental water pollution events.
Year: 2013