This paper presents design technique for a flow field estimator in the sense of LSE(least square estimation). In order for taking into account the strongly nonlinearity of the flow field model, the dimension of the estimator is increased by introducin...
This paper presents design technique for a flow field estimator in the sense of LSE(least square estimation). In order for taking into account the strongly nonlinearity of the flow field model, the dimension of the estimator is increased by introducing the recursive form of the estimated states. Database normalization by standard deviation is also used to enhance the condition number of the covariance matrix. The proposed multi-variable recursive estimator for flow state estimation is promising for the real time implementation compared to the neural network based one. Moreover simulation results show the performance improvements are remarkable than the least square estimation.