Abstract:The measurement acceleration of the forcebalance accelerometer (FBA) is calculated by the control force when the feedback control force is balanced with the external inertia force. Thus, the forcebalance control algorithm is the core of a forcebalance sensor. Most of the traditional control algorithms aim at minimizing the offset position of the sensitive elements from the equilibrium, which limits the measurement accuracy and applicable bandwidth of the forcebalance accelerometer. In this paper, taking a MEMS forcebalance sensor as the object, an optimal control algorithm of forcebalance accelerometer is proposed for minimizing the measurement error. By introducing measurement error as a new state variable, the difficult forcebalance control is transformed into an optimal control problem for response minimization, from which the analytical expression of the optimal control force is obtained. Based on the proposed control strategy, the realtime highprecision detection of unknown acceleration signal is realized. Numerical simulations are carried out for three different types of input acceleration signals (step, periodic and random). It is found that the proposed algorithm can accurately detect all kinds of input acceleration signals, and the frequency band of the measured signal reaches up to kHz. At the same time, the vibration response of the sensitive element can be effectively controlled, which guarantees the large dynamic range of a FBA. Our work provides the basis for the research of high performance forcebalance accelerometer with highprecision and wide frequency band.