基于遗传算法的航空发动机部件特性修正

研究了航空发动机部件特性修正技术，提出了一种基于变适应度函数的模型优化算法，以达到减小总体建模误差，提高模型精度的目的.在稳态模型的基础上，对引气系数、总压恢复系数和各部件的特性进行了修正，使修正后的模型输出与实验数据相一致.采用改进遗传算法，对交叉率和变异率进行了非线性自适应调整，并根据误差大小调整适应度加权系数，避免算法陷入局部最优，同时减小最大建模误差.仿真结果表明，修正后各实验参数平均误差从2.420 8%减小到0.321 7%，模型满足稳态误差小于2%的要求. 

Component map correction of aero-engine based on genetic algorithm

The components characteristic modification technology was studied. A model optimization algorithm based on changed fitness function was proposed to reduce the modeling error and improve the model accuracy. Based on the static model, characteristic correction coefficients of every component, air-entraining correction coefficients and pressure recovery correction coefficients were optimized to achieve high matching accuracy of engine model outputs to test data. The improved genetic algorithm with adaptive crossover rate and mutation rate was adopted, and the weighting coefficient in fitness function was adjusted according to the error. Then the trapping in local optimal solution was avoided and the maximum modeling error was reduced. Simulation results show that the average error of test parameters decrease from 2.420 8% to 0.321 7% after modification, and the model meets the requirements that the static error should be less than 2%.