The binaphthol enantiomers separation process using simulation moving bed technology is simulated with the true moving bed approach （TMB）. In order to systematically optimize the process with multiple productive obj...The binaphthol enantiomers separation process using simulation moving bed technology is simulated with the true moving bed approach （TMB）. In order to systematically optimize the process with multiple productive objectives, this article develops a variant of tissue P system （TPS）. Inspired by general tissue P systems, the special TPS has a tissue-like structure with several membranes. The key rules of each membrane are the communication rule and mutation rule. These characteristics contribute to the diversity of the population, the conquest of the multimodal of objective function, and the convergence of algorithm. The results of comparison with a popular algorithm——the non-dominated sorting genetic algorithm 2（NSGA-2） illustrate that the new algorithm has satisfactory performance. Using the algorithm, this study maximizes synchronously several conflicting objectives, purities of different products, and productivity.展开更多
文摘为研究Eclipse治疗计划系统中正常组织目标(Normal Tissue Objective, NTO)优化工具参数,对脊柱转移瘤立体定向放射治疗计划靶区覆盖剂量和靶区外剂量跌落梯度指数(Gradiet Index, GI)的影响,本研究对10例脊柱转移瘤立体定向放射治疗患者设计两种类型计划。第1种添加NTO优化工具但不添加环优化条件,共得到360个计划,为参考计划;第2种添加环优化条件但不添加NTO优化条件,总共10个计划,为标准计划。所有计划最后剂量归一为100%处方剂量覆盖90%靶区体积。比较两种类型计划的靶区剂量适形度指数(Conformity Index, CI)、靶区外剂量跌落梯度指数、靶区最大剂量(Dmax)、实际治疗时间和计划临床合格率。结果表明,当NTO优化工具参数Fall-off为0.5或者1,Priority为500、700和999时,计划的GI值相似(P>0.05),但是小于其他Fall-off和Priority的GI值(P<0.05);Fall-off和Priority对CI值大小和实际治疗时间没有影响(P>0.05);随着Priority增加,计划靶区体积(Planning Target Volume, PTV)Dmax增加(P<0.05)。参考计划的临床合格率最高为70%。与标准计划相比,最合适NTO优化工具参数计划(Priority、Fall-off分别为500和0.5)的GI值更低(3.449 vs 3.655,P=0.000,Z=-5.534),实际治疗时间没有差异(2.65 min vs 2.32 min,P=0.063,Z=0.832),PTV Dmax偏大(132.1%vs 128.8%,P=0.002,Z=4.214),CI偏大(0.911 vs 0.879,P=0.151,Z=0.984)。综上所述,在计划可通过情况下,当NTO优化工具中参数Distance from Target Border设置为0.2 cm, Start Dose设置为91%处方剂量,End Dose设置为49.4%处方剂量,Fall-off和Priority分别设置为0.5和500时,脊柱转移瘤立体定向放疗计划能够实现较好的靶区剂量适形度、更低的剂量梯度指数以及合适的治疗时间。
基金Supported by the National Natural Science Foundation of China （No.60421002）
文摘The binaphthol enantiomers separation process using simulation moving bed technology is simulated with the true moving bed approach （TMB）. In order to systematically optimize the process with multiple productive objectives, this article develops a variant of tissue P system （TPS）. Inspired by general tissue P systems, the special TPS has a tissue-like structure with several membranes. The key rules of each membrane are the communication rule and mutation rule. These characteristics contribute to the diversity of the population, the conquest of the multimodal of objective function, and the convergence of algorithm. The results of comparison with a popular algorithm——the non-dominated sorting genetic algorithm 2（NSGA-2） illustrate that the new algorithm has satisfactory performance. Using the algorithm, this study maximizes synchronously several conflicting objectives, purities of different products, and productivity.
