Optimization of Dynamic and Buckling Behavior of Thin-Walled Composite Cylinder,Supported by Nature-Inspired Agorithms

The paper presents the optimization of stacking sequence (the lamination angles in subsequent composite layers) of the composite cylinder in order to simultaneously maximize the values of the first natural frequency f1 and the first buckling force Pcr. The optimization problem involves either two objective functions or one which combines both problems using a coefficient whose optimal value is also being searched for. The main idea of the paper is the application of two neural network metamodels which substitute very time- and resource-consuming Finite Element (FE) calculations. The metamodels are created separately through a novel iterative procedure, using examples obtained through Finite Element Method (FEM). The metamodels, once ready, are able to assess the values of f1 and Pcr instantly and thus enable the application of nature-inspired Genetic Algorithm (GA) minimization with reasonable calculation times. Obviously, the maxima of f1 and Pcr may be located in different points of the design parameters (i.e., lamination angles) space, the considered optimization task is to find a solution for which both f1 and Pcr simultaneously reach values as close to their maxima as possible. All the investigated optimization examples are repeated several times and basic statistical analysis of the results is presented.     

ED-NM-MO三联法对丹参三七配比的多目标优化研究

[目的]采用ED-NM-MO三联法对经基线等比增减设计的丹参、三七不同配比的药效学数据进行非线性拟合和多目标优化。[方法]对经基线等比增减设计的丹参、三七不同配比的药效学数据,以心肌缺血程度Σ-ST、心肌缺血程度(缺血区左室)等7个分别反映心肌缺血、心脏状态和血流动力学的指标作为待优化的药效目标,进行非线性拟合和多目标优化。[结果]分别得到针对7个药效指标和6个药效指标(不包含血清中心肌钙蛋白)的Pareto最优配比。[结论]ED-NM-MO三联法是一种适合复方特点的优化方法,可以应用于由多饮片多组分多成分复方药物的剂量配比优化。     

基于支配度迁移模型的多目标生物地理学优化算法

针对传统多目标算法在解决MOPs问题时会出现Pareto前沿收敛结果不好、解集分布性不佳的情况,提出了基于支配度迁移模型的多目标生物地理学算法（MOBBO）。新的迁移模型充分利用了Pareto解之间的支配信息,有助于算法进行有效的个体评价和栖息地排序;为了强化算法的收敛效果,提出了基于优选特征库的自适应迁移策略,以便产生携带较好特征的候选解强化搜索能力;同时为了增强算法进化中Pareto解集的分布性,提出了改进的KNN密度估计方法淘汰过密的个体。通过ZDT和DTLZ系列测试函数以及MDI缩合过程的多目标问题优化上的比较,验证了MOBBO算法具有较快的收敛性和较好的分布延展性。     

非支配排序遗传算法多目标优化金莲花水提工艺的研究

目的:采用非支配排序遗传算法多目标优化金莲花水提工艺最优提取条件,并对其效果进行评价。方法:利用英国Glasgow大学软件工程师陈益提供的Matlab2009a外挂SGALAB工具箱beta5完成遗传算法寻优;SPSS13.0软件进行统计分析。结果:经过非支配排序遗传算法(NSGA)多目标优化后,金莲花水提的岀膏率、总黄酮含量平均水平能达到42.38%和6.84%,金莲花水提取工艺的最优提取条件为加水13倍量、浸泡0.84 h、煎煮3次、每次煎煮1.87 h,并且验证试验达到了较好的效果。结论:在保证多个目标都最优的前提下,NSGA搜索的帕累托非劣解比较理想,提供了可供研究者选择的Pareto非劣解方案,为试验设计最优条件选择提供了合理的方法。     

一种求解阻塞流水车间多目标调度的离散差分进化算法

针对流水车间中产品不存在缓冲区的多目标优化问题,研究了阻塞流水车间的最大完工时间和总流程时间的最小化问题,提出了一种多目标离散差分进化（Multi-objective Discrete Differential Evolution,MDDE）算法搜索Pareto最优调度解。MDDE的变异个体通过非支配解或当前解的邻域随机产生,实验个体通过交叉操作产生,而选择过程则设计为一种多目标选择策略。此外,算法还混合了一种基于插入的Pareto局部搜索方法。基于标准测试算例的数值仿真实验表明,MDDE算法获得的非支配解集在Inverted Generational Distance、Set Coverage和Hypervolume性能指标上均有较好的表现。     

银菊解毒口服液大孔树脂纯化工艺研究

目的在指纹图谱的基础上,联用AHP-CRITIC混合加权法、Plackett-Burman设计(PBD)与Box-Behnken设计(BBD),多目标筛选银菊解毒口服液(YJOL)的纯化工艺。方法采用HPLC法建立YJOL HPLC图谱,以6种指标成分绿原酸、蒙花苷、R,S-告依春、哈巴俄苷、补骨脂素、异补骨脂素的回收率和HPLC图谱相似度为指标,筛选10种大孔树脂中最佳型号。AHP-CRITIC混合加权法赋予6种指标成分的回收率和相似度权重,计算综合指标作为评价标准,运用PBD筛选显著性影响因素,结合BBD优选大孔树脂纯化YJOL的最佳工艺参数。结果 HPD-400型大孔树脂对复方中6个指标成分的富集选择性最高。最佳工艺参数为径高比1∶7,上样pH值3.5,上样质量浓度0.18 g/m L,上样量(药材:树脂)0.96g/g,8 BV的77%乙醇洗脱。纯化后6个指标成分的回收率为78%~98%,HPLC指纹图谱相似度大于0.99,成分均衡回收。模型综合评价预测值为94.28%,实验所得综合评价为93.69%,相对偏差0.59%。结论在HPLC图谱的基础上,联用AHP-CRITIC混合加权法、PBD与BBD筛选YJOL纯化工艺是科学可行的,不仅能提高有效成分的纯度,还能保持化学成分的一致性。     

多目标遗传算法优化金荞麦的提取工艺

目的 优选金荞麦的提取工艺。方法 选择乙醇浓度、溶媒倍数、提取时间、提取次数为考察因素,以总黄酮和浸膏得率为考察指标进行正交试验,运用SPSS 19.0软件建立总黄酮和浸膏得率的二次回归模型,采用多目标遗传算法对其进一步优化,确定金荞麦的最优提取工艺。结果 金荞麦的最佳提取条件：加金荞麦药材24倍量的61%乙醇于80℃提取3次,每次1.90 h;总黄酮和浸膏得率分别为11.58%和22.83%,与遗传算法模型预测结果相符。结论 多目标遗传算法优化的效果理想,所得工艺稳定可行,有效成分得率高,可用于金荞麦有效部位的提取。     

Challenges with multi-objective QSAR in drug discovery

Introduction: The complexity in the drug discovery pipeline, in combination with the exponential growth of experimental and computational data, the technological achievements, and the access to large data sets, has led to a continuous evolution and transformation of quantitative structure–activity relationships (QSAR) to compete with the challenges of multi-objective drug discovery.

Areas covered: After a short overview of the multiple objectives involved in drug discovery, this review focuses on definition of the drug-like space and the construction of local and/or global models, platforms and workflows for step-by-step single-objective optimization (SOO) of the different and often conflicting processes. Multi-targeted drug design is a particular case of multi-objective QSAR integrated into the new era of polypharmacology. Multi-objective optimization (MOO), based on desirability functions or Pareto surfaces and its application in QSAR, as an alternative optimization philosophy, is also discussed.

Expert opinion: Access to large databases as well as to software services by means of cloud technology facilitates research for more efficient and safer drugs. QSAR models implemented in web platforms and workflows provide sequential SOO for multiple biological and toxicity end points, while MOO, still restricted to a limited number of objectives, is helpful for multi-target or selectivity design, as well as for model prioritization.     

In vitro evaluation of multi-objective physiological control of the centrifugal blood pump

Left ventricular assist devices (LVADs) have been used as a bridge to transplantation or as destination therapy to treat patients with heart failure (HF). The inability of control strategy to respond automatically to changes in hemodynamic conditions can impact the patients’ quality of life. The developed control system/algorithm consists of a control system that harmoniously adjusts pump speed without additional sensors, considering the patient’s clinical condition and his physical activity. The control system consists of three layers: (a) Actuator speed control; (b) LVAD flow control (FwC); and (c) Fuzzy control system (FzC), with the input variables: heart rate (HR), mean arterial pressure (MAP), minimum pump flow, level of physical activity (data from patient), and clinical condition (data from physician, INTERMACS profile). FzC output is the set point for the second LVAD control schemer (FwC) which in turn adjusts the speed. Pump flow, MAP, and HR are estimated from actuator drive parameters (speed and power). Evaluation of control was performed using a centrifugal blood pump in a hybrid cardiovascular simulator, where the left heart function is the mechanical model and right heart function is the computational model. The control system was able to maintain MAP and cardiac output in the physiological level, even under variation of EF. Apart from this, also the rotational pump speed is adjusted following the simulated clinical condition. No backflow from the aorta in the ventricle occurred through LVAD during tests. The control algorithm results were considered satisfactory for simulations, but it still should be confirmed during in vivo tests.