ABSTRACT The fast development of urban advancement in the past decade requires reasonable and realistic solutions for transport, building infrastructure, natural conditions, and personal satisfaction in smart cities. This paper presents and explores predictive energy consumption models based on data-mining techniques for a smart small-scale steel industry in South Korea. Energy consumption data is collected using IoT based systems and used for prediction. Data used include the lagging and leading current reactive power, the lagging and leading current power factor, carbon dioxide emissions, and load types. Five statistical algorithms are used for energy consumption prediction:(a) General linear regression, (b) Classification and regression trees, (c) Support vector machine with a radial basis kernel, (d) K nearest neighbours, (e) CUBIST. Root mean squared error, Mean absolute error and Coefficient of variation are used to measure the prediction efficiency of the models. The results show that CUBIST model provides best results with lower error values and this model can be used for the development of energy efficient structural design which helps to optimize the energy consumption and policy making in smart cities.

中文翻译：

---

基于数据分析的智慧城市工业建筑能耗预测模型

摘要 过去十年城市进步的快速发展需要智慧城市的交通、基础设施建设、自然条件和个人满意度方面的合理和现实的解决方案。本文介绍并探索了基于数据挖掘技术的韩国智能小规模钢铁行业的预测能源消耗模型。使用基于物联网的系统收集能耗数据并用于预测。使用的数据包括滞后和超前电流无功功率、滞后和超前电流功率因数、二氧化碳排放和负载类型。五种统计算法用于能耗预测：(a) 一般线性回归，(b) 分类和回归树，(c) 具有径向基核的支持向量机，(d) K 个最近邻，(e) CUBIST。均方根误差、平均绝对误差和变异系数用于衡量模型的预测效率。结果表明，CUBIST 模型以较低的误差值提供了最佳结果，该模型可用于开发节能结构设计，有助于优化智慧城市的能源消耗和政策制定。