Rapid automatic thyristor type excitation controller adjustment via region of required quality construction

The aim of the paper is to introduce a new approach for the Regions of Required Quality (RRQ) construction under Synchronous Generators Control Systems analysis and tuning. Artificial Neural Networks (ANNs) was applied to obtain a suitable model of RRQ. ANNs model of desired region's border allows to get more available information on the performance indices' behavior in the vicinity of the border. A neural network model of RRQ's border is accompanied by available information on the performance indices' gradients behavior in the vicinity of a border. In the opposite of traditional approaches mostly based on approximations through preliminary stored experimental data, proposed method exploits ANNs-model as an element of searching procedure using simulation and experimentation. Theoretical results were applied for RRQ construction for 320KW Synchronous Machine supplied with Thyristor type Excitation System with reference to Rapid Automatic Excitation Controller with six control channels. The obtained regions allowed the best combinations of tunings to provide the desired levels for appointed performance indices related with output voltage, frequency, power and system stability.

[1]  Bart Kosko,et al.  Neural networks for signal processing , 1992 .

[2]  M. Kanat Camlibel,et al.  Upper and Lower Bounds for Controllable Subspaces of Networks of Diffusively Coupled Agents , 2014, IEEE Transactions on Automatic Control.

[3]  M. Goli,et al.  ELLIPSOIDAL APPROXIMATION OF THE TOPOGRAPHICAL EFFECTS IN THE EARTH'S GRAVITY FIELD MODELING , 2014 .

[4]  Magdy M. A. Salama,et al.  Adaptive Self-Adequate Microgrids Using Dynamic Boundaries , 2016, IEEE Transactions on Smart Grid.

[5]  Yuting Wang,et al.  Very Short-Term Load Forecasting: Wavelet Neural Networks With Data Pre-Filtering , 2013, IEEE Transactions on Power Systems.

[6]  Guang-Ren Duan,et al.  On Analytical Approximation of the Maximal Invariant Ellipsoids for Linear Systems With Bounded Controls , 2009, IEEE Transactions on Automatic Control.

[7]  D. Gorinevsky.  An approach to parametric nonlinear least square optimization and application to task-level learning control , 1997, IEEE Trans. Autom. Control..

[8]  Yan Zhou,et al.  Robust Decentralized Data Fusion Based on Internal Ellipsoid Approximation , 2008 .

[9]  Antonio C. S. Lima,et al.  Preventive Control Stability Via Neural Network Sensitivity , 2014, IEEE Transactions on Power Systems.

[10]  Dimitry Gorinevsky,et al.  RBF network feedforward compensation of load disturbance in idle speed control , 1996 .

[11]  Fabrizio Dabbene,et al.  Recursive algorithms for inner ellipsoidal approximation of convex polytopes , 2003, Autom..

[12]  James E. Steck,et al.  Linear and neural network feedback for flight control decoupling , 1996 .

[13]  M. V. Egorchev,et al.  Learning of semi-empirical neural network model of aircraft three-axis rotational motion , 2015, Optical Memory and Neural Networks.

[14]  Mo-Yuen Chow,et al.  A Methodology Using Fuzzy Logic to Optimize Feedforward Artificial Neural Network Configurations , 1994, IEEE Trans. Syst. Man Cybern. Syst..

[15]  Giuseppe Carlo Calafiore,et al.  Approximation of n-dimensional data using spherical and ellipsoidal primitives , 2002, IEEE Trans. Syst. Man Cybern. Part A.

[16]  Marwan Al-Akaidi.  Simulating Speech Coders Using Neural Networks , 1998, Simul..

[17]  Petros A. Ioannou,et al.  Traffic flow modeling and control using artificial neural networks , 1996 .