关键词: 质量交换网络, 连续非结构模型, 过程系统, 优化, 算法

Abstract:

The development of mass exchange networks is a pivotal strategy for minimizing pollutant emissions and reducing economic costs in chemical processes. This study introduced a novel optimization model based on Mixed Integer Nonlinear Programming (MINLP), facilitating the synchronous optimization of mass exchange networks. The model incorporated a continuous non-structured model (NSM), which allowed for the integration of split-stream mixing of unequal concentrations and the consideration of stream stock regeneration. The NSM was characterized by its absence of predetermined structural parameters and fixed matching patterns, using continuous real number intervals to delineate different stream stocks and randomly generating specific real numbers to represent the positions of mass transfer units in these streams. This approach endowed the mass exchange network with remarkable randomness and flexibility, significantly broadening the search domain for optimal solutions. Building on this, the random walk algorithm with compulsive evolution (RWCE) algorithm was employed to simultaneously optimize both integer and continuous variables within the NSM. This algorithm was distinguished by the independent evolution of individual variables. The effectiveness of this model and algorithm was validated through three case studies, which demonstrated that the NSM combined with heuristic algorithms can achieve better results than the current best reported in literature, flexibly representing complex mass exchange network structures and confirming the method’s efficacy.

Key words: mass exchange network, continuous non-structured model, process system, optimization, algorithm