论文部分内容阅读
为提高待生催化剂碳含量预测的准确性,提出一种基于改进的教学算法(MTLBO)来优化BP神经网络的预测模型.针对基础教学算法全局搜索能力差的问题,在教师阶段前后增加了预习和复习过程,并在学生阶段采用量子方式进行更新.测试结果表明,该改进能够提高教学算法全局探索和局部改良能力,利用改进教学算法可优化BP神经网络的权值和阈值,并进行待生催化剂碳含量预测.仿真结果表明,改进后预测模型的预测精度和泛化能力均有一定程度的提高.
In order to improve the accuracy of predicting the carbon content of spent catalyst, a prediction model based on improved teaching algorithm (MTLBO) is proposed to optimize BP neural network.Aiming at the problem of poor global search ability of basic teaching algorithm, pre- And the process of reviewing, and using the quantum way in the student stage to update.The test results show that the improvement can improve the global exploration and local improvement ability of teaching algorithm, use the improved teaching algorithm to optimize the weights and thresholds of BP neural network, Catalyst carbon content prediction.The simulation results show that the prediction accuracy and generalization ability of the improved prediction model are improved to a certain extent.