改进的数据驱动子空间算法求解钢铁企业能源预测问题

本文以钢铁企业生产与能源系统作为研究背景,设计一种数据驱动的子空间方法(data-driven subspace,DDS)预测各生产工序的能源消耗.针对钢铁生产中能源消耗和回收的特点进行了分析,以提取子空间方法的建模因素;为了设计有效的求解方法,对实际生产和数据的特征进行了分析.为了提高预测准确率,文中引入了反馈因子和遗忘因子来改进子空间方法,因子的取值采用粒子群算法(particle swarm optimization,PSO)来优化.对实际生产数据的测试验证了本文所提出的方法的有效性,该结果能够为钢铁企业的能源预测和管理提供有效的决策支持.

Improved data-driven subspace algorithm for energy prediction in iron and steel industry

Using the production and energy system in Iron and Steel industry as the research background, we propose a data-driven subspace (DDS) method for predicting the energy consumption of production operations. The characteristics of energy consumption and regeneration are fully investigated to find the crucial factors for building the model. The features of practical conditions and data are analyzed in designing the efficient solving method. The subspace method is improved by introducing the feedback factor and the forgetting factor, values of which are optimized by particle swarm optimization (PSO) algorithm in order to improve the prediction accuracy. The performance of the improved method is demonstrated by experimental tests using the practical data, which provide beneficial results in energy prediction and management.