多模态复杂工业过程的建模与优化研究

发布时间：2018-07-18 16:04

【摘要】：由于生产工艺的要求,生产过程往往会存在多个不同的稳定工作模态,此为多模态生产过程,过程中各模态的特性各不相同,各模态之间的频繁切换。一般的复杂工业过程建模和优化方法难以适用于多模态过程。研究多模态工业过程的建模与优化方法,对提高多模态工业过程的生产稳定性和生产效率具有重要的意义。在多模态生产过程中,各模态存在各自的特性。过渡模态与稳定模态的操作数据并不存在严格的非此即彼的属性,不同模态对应的操作变量数据的空间密度分布有明显的不同,根据这些特性,选取基于模糊集合理论的模糊C.均值聚类方法,并对该方法引入加权密度进行改进,实现了对多模态数据的模糊聚类分析和模态划分,为多模态过程的建模与优化的研究奠定了基础。利用RBF-ARX模型对划分出的不同模型进行单独建模,通过仿真分析,得到该建模方法对稳定模态和过渡模态的建模均有良好的精度和稳定性的结论。通过对过渡过程的分析,将过渡过程总结为过渡时间确定和不确定这两类,针对存在这两类不同的过渡模态的多模态过程进行优化研究。根据过渡过程关注的过渡时间和资源消耗问题,利用代价函数的思想构建了多模态过程优化的目标函数,考虑生产设备的运行极值和调节能力,设定优化模型的约束条件,引入迭代规划与遗传算法相结合的迭代遗传算法,对存在过渡时间确定的多模态过程进行操作变量的优化,基于双层优化的思想,改进了迭代遗传算法,构建求解存在过渡时间不确定过渡过程的多模态过程优化问题的双层迭代遗传算法。通过仿真研究验证了该优化方法的有效性。
[Abstract]:Because of the requirement of the production process, there are many different stable working modes in the production process, which is the multi-modal production process. The characteristics of each mode are different, and the frequent switching among the modes. General modeling and optimization methods for complex industrial processes are difficult to be applied to multimodal processes. It is of great significance to study the modeling and optimization methods of multimodal industrial processes to improve the production stability and production efficiency of multimodal industrial processes. In the process of multimodal production, each mode has its own characteristics. The operation data of transition mode and stable mode do not have strict attributes of either or, and the spatial density distribution of operation variable data corresponding to different modes is obviously different. According to these characteristics, fuzzy C based on fuzzy set theory is selected. The mean value clustering method is introduced to improve the weighted density, and the fuzzy clustering analysis and modal partition of multi-modal data are realized, which lays a foundation for the modeling and optimization of multi-modal process. The RBF-ARX model is used to model the different models separately. Through simulation and analysis, it is concluded that the modeling method has good accuracy and stability for the modeling of stable mode and transition mode. Through the analysis of the transition process, the transition process is summed up as transition time determination and uncertainty, and the optimization of multi-modal processes with these two different transition modes is studied. According to the problem of transition time and resource consumption, the objective function of multimodal process optimization is constructed by using the idea of cost function. Considering the operating extremum and adjusting ability of production equipment, the constraint conditions of the optimization model are set up. The iterative genetic algorithm combined with iterative programming and genetic algorithm is introduced to optimize the operation variables of multi-modal processes with transition time. Based on the idea of bilevel optimization, the iterative genetic algorithm is improved. A bilevel iterative genetic algorithm for solving multimodal process optimization problems with uncertain transition time is proposed. The effectiveness of the optimization method is verified by simulation.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TB49

【相似文献】