基于机器学习的城市用水量预测模型研究

Research on Urban Water Consumption Prediction Model Based on Machine Learning

随着浙江省水利数字化改革的深入推进,提升水利智慧化管理水平过程中,人工智能、深度学习等技术具有广阔的应用前景。其中运用机器学习算法挖掘水资源动态监测数据的内在规律进行城市用水量动态预测,从而精准掌握城市用水量未来时段的变化情况,对于支撑城市供水旱情研判、提高水资源精细化管理水平具有重要作用。利用长兴水务公司2013—2021年取水实时监测数据,采用支持向量机模型方法构建长兴县城市用水量预测模型,探讨基于机器学习的用水量数据挖掘方法在城市用水量预测领域的适用性,为水资源数字化改革提供有效的模型组件。

With the deepening of the digital reform of water conservancy in Zhejiang Province,using artificial intelligence,in-depth learning and other means to improve the intelligent management level of water conservancy has broad application prospects.Machine learning algorithm is used to mine the internal law of water resources dynamic monitoring data and predict the urban water consumption dynamically,so as to accurately grasp the changes of urban water consumption in the future,which plays an important role in supporting the research and judgment of urban water supply drought and improving the fine management level of water resources.Using the real-time monitoring data of water intake of Changxing water company from 2013 to 2021,this paper constructs the urban water consumption prediction model of Changxing County by using the support vector machine model method,discusses the applicability of the water consumption data mining method based on machine learning in the field of urban water consumption prediction,and provides effective model components for the digital reform of water resources.