Контури керування і обмеження для побудови системи автоматичного керування авіаційним газотурбінним двигуном

Abstract

The choice of programs and control algorithms for ACS determines the possibility of implementation in the engine of the expectancy in the construction of thermodynamic and strength characteristics. For the set modes of operation of the engine limit values of frequency of rotation of rotors of gas temperature , air pressure behind the compressor are defined; on transient modes the size and duration of exceedances by parameters of their maximum admissible values on the set operating modes is limited. The parameters of the GTE adjustment should be chosen so that they characterize the optimality and stability of the engine, as well as the level of loads on its construction. This article considers the process of construction of ACS GTE using control circuits and limitations: free turbine speed support circuit; gas temperature limiting circuit; the circuit of limitations of the resulting frequency of rotation of a rotor of the turbocharger; the circuit of limitations of "physical" frequency of turbocharger rotor rotation, etc. During the operation of each circuit, the value of the derived speed of the turbocharger rotor (minimum and maximum value) is calculated and at the same time the limitation of the fuel consumption is performed. Thus, for the gas temperature limitation circuit, the change in the coefficients allows to achieve a different nature of the temperature change at the throttle response. The value of the coefficient is limited by the value *, which provides a sufficient factor of stability of the regulator. The value was determined by the stable operation of the controller and the presence of the maximum possible delay in the full control circuit . A model with certain transfer functions in the VisSim software environment was constructed for ACS GTE in terms of speed. Measures were taken to stabilize the system: to reduce the gain of the amplifier; changing the parameters of the amplifier and the feedback link; input to the block diagram of the P-regulator system; adding a PI controller system to the circuit; adjustment of the fuzzy regulator; neuro-fuzzy learning. The quality of the system is assessed. A satisfactory reaction time of 2.36 s and an overregulation of 2 % were obtained. The astaticity of the resulting system allows you to accurately maintain the speed in static mode, which means that the system can be used where the accuracy of parameter support is essential.