| Popis: |
In the framework of this work, the question of rationalization of the identification of linear dynamic objects in experimental studies is considered. The current trends in the field of technical system identification techniques are considered. As a starting point for the current study, a previously developed technique for improving the accuracy of identification by cutting off the input signal frequencies was adopted. However, in this case, an application of the obtained results for the technique is considered, which makes it possible to significantly simplify the design of the experiment and the requirements for the test equipment while maintaining the accuracy of identification of dynamic objects. The search for unknown coefficients of the transfer function of the system model is carried out by minimizing the measure of proximity of the pseudo-experimental hodograph of the system and the hodograph of the system model based on regression analysis. As a result, the solution of the identification problem was reduced to solving a system of linear equations. To obtain pseudo-experimental data, a second-order linear system with superimposed additive white Gaussian noise was modeled. An illustrative numerical experiment showed that the error in determining the coefficients of the transfer function does not exceed the range of measurement error of the experimental readings of the hodograph of the system. Further, when using the proposed technique of selecting experimental reference points, it was shown on a similar system that the accuracy of identifying dynamic coefficients can be maintained by using the frequency cutting off technique with a smaller number of reference points. Based on this, an algorithm was proposed for reducing the sample of reference points while maintaining the accuracy of identifying the coefficients of the transfer function at the level of measurement error values. This result was also obtained in an illustrative numerical experiment. The technique described in the publication can be applied in determining the parameters of various types of dynamic objects of complex technical systems in the composition of aviation and space technology.In the framework of this work, the question of rationalization of the identification of linear dynamic objects in experimental studies is considered. The current trends in the field of technical system identification techniques are considered. As a starting point for the current study, a previously developed technique for improving the accuracy of identification by cutting off the input signal frequencies was adopted. However, in this case, an application of the obtained results for the technique is considered, which makes it possible to significantly simplify the design of the experiment and the requirements for the test equipment while maintaining the accuracy of identification of dynamic objects. The search for unknown coefficients of the transfer function of the system model is carried out by minimizing the measure of proximity of the pseudo-experimental hodograph of the system and the hodograph of the system model based on regression analysis. As a result, the solution of the identification problem w... |
