摘要
针对目前压裂泵泵效低、压力波动大,不能满足压裂工艺发展要求的现状,以提高泵效和降低压力波动为优化目标,通过BP神经网络的自学习建立起了压裂泵优化目标与设计变量之间的神经网络模型,并采用VB编程调用Matlab强大的神经网络算法库对BP神经网络模型进行了求解,得到了设计变量的优化值,实现了对700型压裂泵主要结构参数的优化设计。
Aim at the status that low pump effect and large pressure fluctuation of fracturing pump,which can not satisfied with the require of development of fracturing technique.The paper constructs neural model between optimal aim and design variables of fracturing pump by BP neural network in order to raising pump effect and falling pressure fluctuation.Then the BP neural network model is solved through transfer neural network algorithm warehouse of Matlab by VB program.At last,the optimal design of main structural variables is realized and the optimal parameter is get.
出处
《机械》
2007年第9期48-50,共3页
Machinery
关键词
压裂泵
BP神经网络
优化设计
fracturing pump
BP neural network
optimal design
