论文部分内容阅读
根据大量含Re镍基单晶高温合金试验所测TCP相含量的数据,采用一种先进的人工神经网络方法建立运算模型,对合金的TCP相含量进行预测,并与报道所用回归方法进行了比较。结果表明,所建网络模型能够比较准确地预测含Re镍基单晶高温合金中TCP相的含量。将正交分析与网络预测相结合,获得几种主要合金元素对TCP相含量的影响顺序。
According to the data of TCP phase content measured in a large number of Re-containing nickel base single crystal superalloy test, an advanced artificial neural network method was used to establish the operation model to predict the TCP phase content of the alloy and compare with the reported regression method . The results show that the proposed network model can accurately predict the content of TCP phase in the Re-containing nickel base single crystal superalloy. Orthogonal analysis and network prediction are combined to obtain the influence order of several main alloying elements on the TCP phase content.