Stochastic model and optimum sampling interval of variables for environmental measurement systems

Abstract

It is common in engineering to model time-dependent variables as diffusion process represented by stochastic differential equations. This is usually helpful when empirical datasets describing time evolution of variables are available. This helps in accurate estimation of parameters of the stochastic differential equation which describes the dynamic system. Additionally, it helps in characterization and determination of optimal performance of the system. The above have been conducted in this study using real environmental field data. Linear stochastic model was fitted to longitudinal datasets and optimum sampling interval investigated. A new method has been proposed for determination of optimum sampling interval. Results obtained differ from those of hypothetical optimum which do not take energy consumption into consideration.