论文部分内容阅读
介绍了离散灰色理论的建模机理,提出Levenberg-Marquardt算法改进的B P神经网络,分析两种预测模型单独使用时的优缺点。将离散灰色模型与改进的BP神经网络进行有机结合,给出利用BP神经网络对离散灰色模型残差进行修正的综合预测方法。接着利用该预测模型对大连储备库某罐体不均匀沉降进行预测。结果表明,综合预测模型弥补了单一预测模型的缺点具有更高的预测精度。
The modeling mechanism of discrete gray theory is introduced. An improved BP neural network based on Levenberg-Marquardt algorithm is proposed. The advantages and disadvantages of the two prediction models are analyzed. The discrete gray model and the improved BP neural network are combined organically, and a synthetic forecasting method based on the BP neural network to correct the residual of the discrete gray model is given. Then, the prediction model is used to predict the uneven settlement of a tank in Dalian Repository. The results show that the comprehensive prediction model makes up for the shortcomings of a single prediction model and has higher prediction accuracy.