: ..,
..,
..,
..
:
: 119
:
: 2026
: .., .., .., // . - 2026. - . 119. - .284-297.
: , , , , ,
(.): gas turbine engine, overdetermined object, selector, adaptive control, fuzzy controller, threshold model of collective behavior
: () (). , , ( ) ( , .). , , . : , , . , . , , (), . (). , . , . , MATLAB Simulink, , . .
(.): Improving the efficiency and safety of aircraft is directly related to improving the automatic control systems (ACS) of their gas turbine engines (GTE). The key problem here is that the GTE is an overdetermined object, where the number of control actions (fuel consumption) is less than the number of regulated output parameters (rotation speed, temperature, etc.). The traditional solution to this problem is to use a selector - a functional element that selects one, the most critical control channel at any given time. However, this approach has its drawbacks: abrupt switching between channels, the phenomenon of "chatter", deterioration in the quality of transient processes and the impossibility of taking into account the requirements of all circuits simultaneously. The article provides a comparative analysis of the evolution of methods aimed at overcoming the shortcomings of selector control. The adaptive selector with additional signal self-tuning loops for coordinating switching moments and an adaptive fuzzy group controller (AFGC) replacing discrete logic with fuzzy logic are considered. The threshold model of collective behavior (TMCB) is proposed as a fundamentally new approach. This model interprets each controlled parameter as an autonomous agent making a binary decision based on the influence of neighboring agents and its own threshold. This allows moving from selection to cooperative interaction, while simultaneously taking into account the requirements of all control channels. The results of modeling performed in the MATLAB Simulink environment demonstrate the advantages of TMCB over traditional methods in such quality indicators as overshoot and transient time. A conclusion is made about the prospects of using multi-agent threshold models for constructing intelligent group controllers of gas turbine engines
PDF
: 27, : 19, : 17.