摘要
应用改进的BP神经网络算法对过渡区内流量变化下的蒸汽管路摩阻系数进行了预测,解决了因蒸汽流量变化而使得摩阻系数值计算过程复杂繁琐、效率低下的问题。实例表明,该方法得到的仿真值与Colebrook方程计算值符合较好,所产生的误差平均水平能够满足工程计算的精度要求。
An improved BP neural network algorithm was used to predict the friction factor in transition region under variable flow of steam pipeline,the problem of the calculating process complicated and low efficiency for calculating the friction factor when the steam is variable was solved.A given example shows that the simulated results agree well with the calculated results obtained by the Colebrook formula,and Calculation errors were in the range of engineering precision.
出处
《计算机与数字工程》
2010年第10期22-24,27,共4页
Computer & Digital Engineering
