Instruments and Systems: Monitoring, Control, and Diagnostics Annotation << Back The construction of functional dependencies based on empirical data recovery problem of altitude and speed and throttle characteristics of the turbojet Veresnikov G.S. IKRYANOV I.I. The article discusses approaches to solving the problem of recovering the altitude and speed and throttle characteristics of the turbojet engine. Solution of this problem required for calculating the aerodynamic, which is an important step in the development of new aircraft. The desired range of the original data is significantly wider than available, so in this paper we solve the problem of expanding the range of powerplant data by their interpolation and extrapolation. As a result of experiments had identified the need for preprocessing of the input data to produce satisfactory results. The article describes the experimental data for solving two arising in practice task, using multiple nonlinear regression techniques, cubic splines and neural networks. According to the research results is a comparison of methods are distinguished advantages and disadvantages. Keywords: characteristics of the turbojet engine, neural networks, multiple nonlinear regression, cubic splines, interpolation, extrapolation Contacts: E-mail: veresnikov@mail.ru Pp. 26-36.