Parameter Estimation of the Solid-State Wave Gyroscope on the Basis of the Neural Network Autoregression Algorithm for Time Series Prognosis

The method for solving the problem of identifying the parameters of a solid-state wave gyroscope from the data of time series of signals received from the information pickup sensors is proposed. The dynamic model of the gyroscope is described by a system of second-order ordinary differential equations with unknown coefficients associated with the device parameters (nominal eigen frequency of oscillations, nominal decay factor, amplitudes and angles of the 4th harmonic of the inhomogeneity with respect to mass and rigidity). The parameter identification algorithm is based on the use of a single-layer autoregression neural network (NNAR) which implements the forecast of the dynamics after learning. The found weight coefficients of the neural network make it possible to identify the required parameters of the gyroscope resonator.