遗传规划在电化学法检测转子钢热脆性中的应用  

Genetic Programming Approach to Predicting Temper Embrittlement of Rotor Steels Using Electrochemistry Technique

作　　者：张胜寒[1] 范永哲[1] 陈颖敏[1] 李向阳[2] 

机构地区：[1]华北电力大学环境科学与工程学院,河北保定071003 [2]钢铁研究总院,北京100081

出　　处：《钢铁》2006年第5期69-72,77,共5页Iron and Steel

基　　金：国家电力公司资助项目:汽轮机转子钢热脆化非破坏性检测法的开发(SP11-2001-02-29)

摘　　要：在汽轮机转子钢热脆性电化学法无损检测技术的开发中,将遗传规划法应用到30Cr2MoV钢脆性转变温度预测模型的建立中,以提高其检测精度。把材料的脆性转变温度作为预测模型的因变量,将动电位再活化法测得的二次活化峰电流密度、电解液温度、材料的化学成分参数(J参数)、材料中Cr含量和晶粒度作为自变量。将因变量和自变量的试验测定结果分为训练样本数据和检验样本数据。依据训练样本数据,经过遗传规划法求得最优预测模型,并用检验样本数据对所得的预测模型进行检验。结果表明,所得预测模型的预测误差为±20℃,模型精度比采用传统的多元线性回归法得到的模型高1倍多。因此,可以将遗传规划法应用到汽轮机转子钢预测模型的建立中。The genetic programming approach was proposed to predict temper embrittlement of rotor steel （30Cr2MoV）. Two independent data sets were obtained experimentally： training data and verifying data. Peak current density of reactivation measured by the potentiodynamic anode polarization method, temperature of electrolyte, the chemical composition of steel （J-factor）, Cr content and the grain size of steel were used as independent variables, while fracture appearance transition temperature as dependent variable. On the basis of training data, the optimum model was obtained by genetic programming, and the accuracy of it was verified with the verifying data. The prediction error of the model is within ±20℃. The accuracy of this model is better than that of the model obtained using multiple linear regression method. The results suggest that, the model obtained by genetic programming is feasible and effective.

关 键 词：汽轮机转子 热脆性 电化学极化法 遗传规划 

分 类 号：TG115[金属学及工艺—物理冶金]