论文部分内容阅读
应用均匀实验设计和支持向量机方法构建复杂过程系统的经验模型(元模型),并将其作为适应度函数与遗传算法结合,建立了该系统的优化方法.该方法只需采用少量仿真模型计算数据便可建立复杂过程系统的元模型,可显著降低复杂过程系统模型的计算过程,便于复杂过程系统的优化.将该方法用于普光高含硫天然气净化装置全流程操作参数优化,在操作参数优化空间内均匀选取10个实验点,建立了净化装置全流程元模型,其预测值的相对误差小于4%.优化结果表明,在优化操作点,净化装置有效能效率提高了6.6%.
An empirical model (meta-model) of a complex process system is constructed by means of uniform experimental design and support vector machine method, which is used as a fitness function and genetic algorithm to establish an optimization method of the system. The proposed method requires only a few simulation models Data can be used to establish the metamodel of complex process system, which can significantly reduce the calculation process of complex process system model and facilitate the optimization of complex process system. This method is used to optimize the operating parameters of the whole process of Puguang high sulfur natural gas purification plant. The 10 experimental points are uniformly selected in the optimization space, and a full-process meta-model of the purifying device is established, the relative error of the predicted value is less than 4% .The optimization results show that the effective energy efficiency of the purifying device increases by 6.6% at the optimized operating point.