Page 232 - Neural Network Modeling and Identification of Dynamical Systems
P. 232

6.4 SEMIEMPIRICAL MODELING OF AIRCRAFT LONGITUDINAL TRANSLATIONAL AND ANGULAR MOTION  223






















































                          FIGURE 6.14 The coefficient C z (α,φ) for various values of φ according to (A) the data from [20] and (B) the error of its
                                        for fixed q = 0 deg/sec and V = 150 m/sec (From [38], used with permission from Moscow Aviation
                          approximation E C z
                          Institute).

                          ter 2. These algorithms were implemented using  form. The Jacobi matrix is calculated using the
                          the Matlab Neural Network Toolbox and rely on  RTRL algorithm [22].
                          the Levenberg–Marquardt algorithm for min-     ANN modules for functions C x , C z ,and
                          imization of the mean square error objective  C m are represented by sigmoidal feedforward
                          function for the neural network in the LDDN  ANNs. The variables α, δ e ,and q/V are used as
   227   228   229   230   231   232   233   234   235   236   237