摘要
针对工业用水量的特点,建立了改进的BP神经网络用水量预测模型,采用遗传算法对BP神经网络权系进行优化改进,改进的BP神经网络算法预测结果好于灰色理论预测和BP算法预测。以本溪市某供水厂用水量数据对改进的BP神经网络模型进行训练并预测,将其预测结果与灰色理论预测和BP神经网络预测结果进行比较分析,得出该方法用于供水系统用水量预测误差较小,同时克服了其他两种算法的缺陷。
Based on the characteristics of water consumption, the paper established the model to forecast water consumption with the improved BP neural network and applied genetic algorithms to optimize the weight matrix. The forecast result shows that the improved BP neural network is better than the one only using the Grey theory forecasting or BP neural network forecasting. The model is trained with the water consumption data of Benxi Iron-steel Company and used to forecast water consumption. Comparing the forecast results with the Grey theory forecasting and BP algorithm it is concluded that the improved BP neural network model has the small error in forecasting water consumption, at the same time it can overcome shortcomings of other algorithms.
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2004年第2期191-193,共3页
Journal of Liaoning Technical University (Natural Science)
基金
辽宁省教育厅攻关计划基金资助项目(202183392)
