Abstract

Clay drying is one of the most important steps in the construction materials industry. Due to its high-energy requirement, clay drying is one of the most energy-consuming processes contributing to greenhouse gas emissions, motivating the search for strategies to increase the energy efficiency of the process. The process involves several operating variables whose historical records can generate a large amount of data, which can be organized and analyzed efficiently using data mining techniques (DMT). In this paper, the performance variables of a material-industry rotary dryer were identified and analyzed. A cluster analysis was performed using the Ward’s classification method with 1729 operating data, identifying the kind of materials processed and the operating configurations. A linear regression model was constructed using the RRELIEFF feature selection algorithm in Orange: Data Mining Toolbox in Python. The main thermal-energy-consumption variables were the production rate, power, inlet gas temperature, and product moisture delta. These four variables represent the dryer’s specific energy consumption index ($I_c$) more suitably than the classic Energy Vs Production model. Consequently, it was established that a global-aeration-factor reduction could improve the process efficiency by 7% to 10% at the average production rate.

摘要

粘土干燥是建筑材料行业中最重要的步骤之一。由于其高能量需求，粘土干燥是导致温室气体排放的最耗能过程之一，促使人们寻求提高过程能源效率的策略。该过程涉及几个操作变量，其历史记录可以生成大量数据，可以使用数据挖掘技术 (DMT) 有效地组织和分析这些数据。在本文中，对材料工业旋转干燥机的性能变量进行了识别和分析。使用 Ward 分类方法对 1729 个操作数据进行聚类分析，确定处理的材料种类和操作配置。使用 Orange: Data Mining Toolbox in Python 中的 RRELIEFF 特征选择算法构建线性回归模型。主要的热能消耗变量是生产率、功率、入口气体温度和产品水分增量。这四个变量代表干燥机的比能耗指数（${我}_C$) 比经典的能源与生产模型更合适。因此，确定了全球曝气系数的降低可以在平均生产率下将工艺效率提高 7% 到 10%。