Performance Optimization Of Variable Cycle Engine Based On Intelligent Optimization Algorithm

Variable cycle engine(VCE)combines the advantages of turbofan engine and turbojet engine in performance,as the VCE has a greater control difficulty because of the characteristics of multi-variable,multi-constrained,highly nonlinear,highly coupled between the loops,it is necessary to design a more advanced control system.In this paper,the performance optimization control of variable cycle engine based on intelligent optimization algorithm is studied.Teaching-Learning-Based Optimization Algorithm(TLBO)is widely used in engineering problem optimization because of its simple structure,few parameters and strong search ability.In this paper,the basic principle and convergence of TLBO algorithm is studied,and it provides the theoretical guidance for the improvement of the algorithm.In order to reduce the possibility of falling into local optimum,an improved algorithm of TLBO(MTLBO)is proposed.Including the following aspects: In the initialization phase,"reverse learning" is introduced to increase the diversity of the population;in the teaching phase,the "inertia weight" of the particle swarm optimization algorithm is introduced,and it guides the students' learning direction and balances the global and local search capability;in the learning phase,the two learn from each other is improved into two or three mutual learning in order to enhance the diversity of the search space;In the later iteration of the algorithm,the small range Gaussian perturbation is added to enhance the dispersion of the search.The MATLAB simulation based on MTLBO is completed by using the BenchMark test function,compared with the Optimization results of PSO,ABC and basic TLBO algorithm.The result shows that MTLBO has better convergence precision and convergence speed in the optimization of low-dimensional and high-dimensional BenchMark test functions.This paper designs and implements a multivariable performance optimization control algorithm based on MTLBO.In the double-culvert and single-culvert work mode,it can complete the optimization of the maximum thrust and minimum fuel consumption performance of the variable cycle engine,and the influence of TF on the optimization results is analyzed.At the same time,it is shown that the MTLBO algorithm has better optimization effect than the genetic algorithm on the maximum thrust optimization of the variable cycle engine in the single thrust mode with nonafterburner model.