Abstract
Enabled by the increased availability of data, the data assimilation technique, which incorporates measured observations into a dynamical system model to produce a time sequence of estimated system states, gains popularity. The main reason is that it can produce more accurate estimation results than using either a simulation model or the measurements. Due to this benefit, the data assimilation technique has been applied in many continuous systems applications, but very little data assimilation research has been found for discrete event simulations. With the application of new sensor technologies and communication solutions, the availability of data for discrete event systems has increased as well. The increased data availability for discrete event systems but the lack of related data assimilation techniques thus motivated this work on data assimilation for discrete event simulations.
Since discrete event simulations are highly nonlinear, nonGaussian systems, particle filters are used to conduct data assimilation in discrete event simulations. However, applying particle filtering in discrete event simulations still encounters several theoretical and practical problems, such as the state retrieval problem (discrete event simulation models have a piecewise constant state trajectory, so the retrieved state was updated at a past time instant, with which inaccurate estimation results will be obtained), the variable dimension problem (the dimension of the state trajectory during a fixed time interval is random, leading to inapplicability of the standard sequential importance sampling algorithm), and the processing of nonnumerical data. Therefore, this research aims to develop a particle filter based data assimilation framework for discrete event simulations, in which the aforementioned problems can be addressed.
Since discrete event simulations are highly nonlinear, nonGaussian systems, particle filters are used to conduct data assimilation in discrete event simulations. However, applying particle filtering in discrete event simulations still encounters several theoretical and practical problems, such as the state retrieval problem (discrete event simulation models have a piecewise constant state trajectory, so the retrieved state was updated at a past time instant, with which inaccurate estimation results will be obtained), the variable dimension problem (the dimension of the state trajectory during a fixed time interval is random, leading to inapplicability of the standard sequential importance sampling algorithm), and the processing of nonnumerical data. Therefore, this research aims to develop a particle filter based data assimilation framework for discrete event simulations, in which the aforementioned problems can be addressed.
Original language  English 

Awarding Institution 

Supervisors/Advisors 

Award date  27 Feb 2018 
Print ISBNs  9789461868930 
DOIs  
Publication status  Published  2018 