摘要
针对BP神经网络在训练过程中存在局部最优解和收敛速度慢的不足,将具有良好全局搜索能力的遗传算法应用到BP神经网络的训练过程,建立了GA-BP神经网络模型,并将其应用到板坯连铸漏钢预报系统中;结合某钢厂板坯连铸现场历史数据对该模型进行了测试,测试结果以97.56%的预报率及100%的报出率证明了GA-BP神经网络漏钢预报方案的可行性。
Aiming at the two terrible drawbacks of slow convergence and local optimal solution in the training process of BP neural network, genetic algorithm was introduced to the training process of the BP neural network to improve its converge property, so a GA-BP neural network was established, and then it was introduced into the breakout prediction system. The GA-BP breakout prediction neural network model was trained and tested with the historical data collected from a steel plant. The results show that the convergence rate of the GA-BP neural network model is significantly improved compared with the traditional BP neural network, and the feasibility of the model is verified by the testing result with the accuracy rate of 97.56% and the prediction rate of 100%.
出处
《铸造技术》
CAS
北大核心
2014年第4期736-739,共4页
Foundry Technology
基金
河北省科学技术研究与发展计划项目(07212119D)
关键词
遗传算法
BP神经网络
连铸
漏钢预报
genetic algorithms
BP neural network
continuous casting
breakout prediction