Artificial neural networks inversion based dynamic compensator of sensors

摘要

A novel dynamic compensator of sensors based on ANN (artificial neural networks) inversion is proposed according to the theory of inverse methods and excellent approach abilities of ANN. ANN inversion-based dynamic compensator of sensor is achieved through offline training the ANN inversion that combines the static ANN and dynamic factors. The dynamic performance of sensor is improved effectively by cascading ANN inversion with the original sensor. Simulation and experiment results indicate that the dynamic performance of sensor is remarkably improved and enhanced using the proposed compensator.