基于GA-BP神经网络的柴油喷雾贯穿距预测

为解决柴油喷雾贯穿距测量的问题,提出一种基于GA-BP神经网络的预测方法。首先通过实验得到30组柴油在定容弹中不同环境背压、喷油压力和喷油脉宽等条件下的喷雾贯穿距,然后将前20组数据作为训练样本,后10组数据作为测试样本,最后分别通过BP神经网络和GA-BP神经网络建立喷雾贯穿距的预测模型。研究结果表明:GA-BP神经网络预测模型的平均相对误差和相对误差方差均比BP神经网络预测模型的低,并且其达到收敛时所需的迭代次数比BP神经网络预测模型的少。基于GA-BP神经网络的柴油喷雾贯穿距预测模型具有较高精度和适用性,为喷雾贯穿距的测量提供了一种低成本、高效率的方法。     

N-Terminal Truncation of TACO Inhibits PMA-Induced U937 Cell Adhesion

The effect of TA001-299, the N-terminal truncation of TACO, on phorbol 12-myristate 13-acetate （PMA）-induced U937 cell adhesion was investigated. Full-length TACO and several truncations were overexpressed in U937 cells. The effects of the expressed proteins on U937 cell adhesion mediated by PMA-induced differentiation were observed by fluorescence microscopy. The results show that the overexpression of TACO1-299 inhibits cell adhesion while overexpressions of the other proteins do not have this effect. The actin-binding capability of TACO1-299 was investigated and the results show that TACO1-299 lacks the ability of TACO to bind F-actin. The inhibitive effect of TACO1-299, the functional domain of TACO, suggests that TACO may play a role in cell differentiation mediating adhesion of monoblastic leukemia cells.