Analysis of complex structures of marine systems with attraction methods of neural systems
Sergei Chernyi, Anton Zhilenkov · 2015
We have given the results of developing of a method of neuro-fuzzy structures selforganization in intelligent process control systems. The proposed modification of the basic algorithm can improve the control performance index of intelligent automated control systems at a reduced volume of calculations and corresponding increase of system performance. In classical training rule, fuzzy neural networks and the number of production rules, type of membership functions, fuzzy inference of algorithm type, etc. is given a priori and does not change during the network training. In the case of an incorrect choice of these parameters fuzzy neural networks can be ineffective in the field of automation. Operation of the developed algorithm is based on the theory of sampling frequency and training frequency distribution. In traditional adaptive control systems parameters are adjusted once every sampling period, thus the sampling rate and update rate are not separated. In order to reduce the algorithm running time and improve its efficiency when performing parametric synthesis of asymptotically stable intelligent control systems, the experts determine method of the concentration coefficient of membership functions and sampling limits for further adjustments to the base of the adaptive-established rules. One of the most essential tasks for a number of systems of the automatic controls in the autonomous electric power systems of water transport is accurate calculation of variable harmonic components in the non-sinusoidal signal. In the autonomous electric power systems being operated with full semiconductor capacity, the forms of line currents and voltages are greatly distorted, and generator devices generate voltage with inconsistent frequency, phase and amplitude. It makes calculation of harmonic composition of the distorted signals be a non-trivial task. The present paper provides a mathematical set for solution of the outlined problem including the realization in the discrete form. The simplicity and efficiency of the system proposed make possible to perform its practical realization with the help of cheap FPGA. The test of the developed system is performed in the Matlab medium.