基于神经网络的高温压力管道涡流检测信号的温度补偿研究

Signals Compensation Research of NN-based Hightemperature Pressure Pipe Eddy Current Testing

压力管道的质量判定可以采用涡流检测的方法,分析了温度对涡流检测的影响,实验测定了一定的数据,设计了神经网络补偿器,采用实验样本数据对神经网络进行了训练,训练后的神经网络能准确地预测温度对检测信号的影响。

Eeddy current testing method can be used to determine pressure pipe quality.The temperature impact on eddy current testing was analyzed and experimental data were obtained,including neural network（NN） compensator design,and training neural network with experimental sample data.The trained neural network can accurately forecast temperature impact on detection signal.