bp人工神经网络在材料失效中的运用.doc

摘要

火电厂锅炉过热器金属管道在常规运行下疲劳失效是影响发电效率的重要因素之一。本文基于热电厂金属过热器管道的失效分析报告中的实验参数，应用matlab的BP人工神经网络模拟功能对失效对已经失效的管道进行寿命检验。同时，通过这样的检验步骤可以说明人工神经网络在此问题上比较可靠（但非精确）。

关键词：MATLAB  BP人工神经网络  高强度合金  失效

Abstract

Failure of the boiler pipes in Thermal power plant is one of the most important factors which deeply impact the efficiency of power generation in the situation of normal operation. Based on the data in the report of metal pipe failure of the power plant, the functions of simulation BP artificial neutral network (ANN) in MATLAB program are applied to verify the life of the pipe which has been broken in this paper. Meanwhile，it can be proved by the step of the experiment that ANN is relatively（but not exactly） accurate in this situation.

Keywords: MATLAB, BP artificial neutral, network，High strength alloy, Failure