Вибір методу оптимізації систем автоматичного керування в системах автоматизованого інжинірингу

Анотація

SELECTION OF METHOD TO OPTIMIZE ACS IN CAE

O. V. Vasylenko, Y. I. Petrenko

The increasing complexity of the tasks and functions of automatic control systems (ACS) and the continuous development of the mathematical basis of programs for their computer-aided design leads to the fact that the choice of the most appropriate method and means of optimum design solution becomes a problem. The task of improving the methodological support of optimization automatic control systems (ACS) on the structural and physical levels of abstraction in systems computer-aided engineering (CAE) was performed in this paper. The object of research was the process of computer-aided design of ACS in the CAE, the subject of investigation - mathematical and methodical support of parametric optimization. Objective: development methodological support of computer-aided design ACS in CAE for the example of program 20-sim, for more efficient using of this tool in the synthesis procedures of optimal design solutions. Among study methods used in this research - simulation and synthesis of the results. Methods of parametric optimization of multidomain ACS in universal mathematical processors and computer-aided engineering programs (CAS and CAE, respectively), were investigated in this work.

Classification of existing methods of optimization, definition of their common features and principles, allowed developing the comparative table and determine the rating, according to which it is recommended to choose the appropriate method in the programs of automated engineering calculations, engineering and design. The parameters optimization depends on the modeling approaches (abstraction level, causal or a-causal approach, singularity of model etc) and is based on the mathematical software of simulation process.

Advantages of tools for ACS synthesis in 20-sim (Controller Design Editor, Linear System Editor and Filter Editor), including the possibility of synthesis and analysis of the blocks with delay are shown. Area of adequate application described above instruments is linear causal models of ACS on the macro-level, which leads to limitations in the study of such systems, in particular, can not be estimated accuracy of reproduction transients. Besides, nonlinear system must be linearized in operating point using frequency domain tools.

According to the analysis of mathematical and software of CAE and ACS simulation by the structural and physical approaches in program 20-sim, identifies areas adequate application of synthesis tools and methods of parametric optimization. By summarizing the results of empirical research, we developed an algorithm and practical recommendations for choosing a method of parametric optimization in 20-sim. The developed algorithm helpful for users who designs ACS in CAS system.

The obtained practical results can be useful to the engineers-developers of automatic control systems and can be used in the educational process.