论文部分内容阅读
通过对Q195钢丝在不同温度、时间下的退火处理,测试了退火前后的抗拉强度。采用BP神经网络建立了Q195钢丝连续退火后抗拉强度与初始抗拉强度、钢丝直径、保温时间和退火温度之间的预测模型,对钢丝连续退火后的抗拉强度进行预测。结果表明:BP网络预测最大相对误差为3.49%。该预测模型对于Q195钢丝连续退火抗拉强度的预测是有效的、可行的。
The tensile strength of Q195 steel before and after annealing was tested by annealing at different temperature and time. The prediction model of tensile strength, initial tensile strength, wire diameter, holding time and annealing temperature after continuous annealing of Q195 steel wire was established by BP neural network. The tensile strength after continuous annealing of wire was predicted. The results show that the maximum relative error of BP neural network prediction is 3.49%. The prediction model is effective and feasible for the prediction of the tensile strength of Q195 continuous annealing wire.