基于改进型遗传算法的打浆优化

Beating Optimization Based on Improved Genetic Algorithm

针对造纸生产线上打浆过程中的打浆优化问题,以基于神经网络的软测量模型辨识为基础,结合改进的遗传算法,解决了标准遗传算法中种群过早收敛和促使种群跳出局部极值点,实现了打浆全局最优控制。仿真结果验证了该算法的可行性与有效性,实践表明这种优化控制方法使打浆能耗明显降低,达到了节能的优化目的。

Focusing on the beating optimization in paper making, the global optimization for beating is realized via neural-network soft sensing model identification and improved genetic algorithm which avoids population converging too early and local maximums in the standard genetic algorithm. Simulation verifies the algorithm is feasible and valid. Practice results show this optimization control method can decrease beating power cost significantly.