Application of the adjusted ordinary least square in the error model Identification of accelerometer

Due to the fact that the input acceleration is a measurement result, contaminated by noise while accelerometer is tested on centrifuge. The statistical character of the result deteriorates by applying ordinary least square method when identifying the model parameter. In order to solve the problem of the input noise, firstly, the limitation of the ordinary least square method is analyzed, secondly, theory of adjust least square of EV model is utilized to identify the error model of accelerometer testing on centrifuge. The input noise is taken into account, then the ordinary least square is adjusted to ensure the estimation unbiased and consistent. The simulation results demonstrate that the proposed method outperforms the ordinary least square method.