Inversion technology of transitional history of negative step-force based on neural networks

In the paper, a kind of new metrology of negative step-force is put forwards based on neural networks. Firstly, the dynamometric system of negative step-force generator is introduced which contains a special design of measuring cell that is connected in series with a force transducer. Secondly, a neural network model between the negative step-force and the displacement response of a certain point on the surface of the cell is built based on the technology of finite element analysis. Then according to the displacement response of the point on the cell, the dynamic force that acts on the transducer can be inversed, and the method proves to be true theoretically. Finally, a kind of laser Doppler interferometer is designed to measure the velocity of the point on the surface of the cell, and the displacement is calculated in the meantime, and then the transitional history of negative step-force could be got based on the neural network model, and the method proves to be correct experimentally.