基于改进遗传算法的燃气轮机自适应建模

综合运用了一种改进的遗传算法和自适应建模技术对燃气轮机的精确特性进行寻优获取.引入自适应机制优化交叉和变异算子,同时引入模拟退火算法,使改进遗传算法能很快接近最优解,并能跳出局部最优的陷阱,在保证解的质量的同时提高了收敛的速度.针对以往自适应模型中未考虑测量参数间的线性相关性和不同的传感器测量精度对目标函数的影响等问题,采用加权方法建立了较为完备的燃气轮机自适应数学模型,应用改进遗传算法获取燃气轮机部件的精确特性,实例计算结果表明:模拟退化改进遗传算法进行的自适应建模效果更好.

Adaptive simulation of gas turbine performance using improved genetic algorithm

The precision performance of gas turbine was obtained by synthetically using an improved genetic algorithm and adaptive modeling techniques.Adaptive mechanisms were also added to optimize the crossover and mutation operators,and simulated annealing was also imported.The improved genetic algorithm can approach the optimal solution quickly,and trip out the trap of locally optimal solution.It can insure the solution quality and improve the convergence speed.Considering that previous studies did not take into account the problems that the objective function was impacted by some measuring parameters which had linear correlation and sensor accuracy class,a more maturity adaptive model was established by using the weighted method,and the algorithm was used to search the precision performance.The results show that the improved GA-based performance adaptation method is more robust.