Blast furnace modeling and control is one of the important problems in the industrial field, and the black-box model is an effective mean to describe the complex blast furnace system. In practice, there are often different learning targets, such as safety and energy saving in industrial applications, depending on the application. For this reason, this paper proposes a framework to design a domain knowledge integrated classification model that yields a classifier for industrial application. Our knowledge incorporated learning scheme allows the users to create a classifier that identifies "important samples" (whose misclassifications can lead to severe consequences) more correctly, while keeping the proper precision of classifying the remaining samples. The effectiveness of the proposed method has been verified by two real blast furnace datasets, which guides the operators to utilize their prior experience for controlling the blast furnace systems better.
翻译:高炉建模和控制是工业界的重要问题,而黑箱模型是描述复杂高炉系统的有效手段之一。在实际应用中,根据不同的应用,往往存在不同的学习目标,例如安全和节能。为此,本文提出了一个框架,用于设计一个融合领域知识的分类模型,产生适用于工业应用的分类器。我们的知识集成学习方案允许用户创建一个分类器,更准确地识别“重要样品”(其错误分类可能会导致严重后果),同时保持正确分类其余样品的适当精度。所提出方法的有效性已通过两个真实高炉数据集得到验证,该方法指导操作员更好地利用其之前的经验来控制高炉系统。