摘要
针对铜闪速熔炼过程中冰铜品位检测的重要性,根据多相多组分数学模型建立冰铜品位的机理模型;同时该过程具有大滞后、非线性等复杂特性,利用现场的大量生产数据建立模糊神经网络模型,并提出一种新的网络参数学习的受约束梯度下降算法,提高其参数学习效率。基于模糊逻辑的智能协调器根据实际生产条件融合两种模型的输出作为预测结果。工业数据验证表明,智能融合模型比单一模型更能有效地实现冰铜品位的准确预测,为铜闪速熔炼过程的优化控制提供有力的指导。
Due to the importance of detecting the copper matte grade in the copper flash smelting process, the mechanism model was established according to the muhiphase and multi-component mathematic model. The fuzzy neural network model was set up through a great deal of production data. A constrained gradient descent algorithm which was used to update the parameters was put forward and the learning efficiency was improved. An intelligent coordinator based on fuzzy logic is proposed to synthesize the output of two models as the prediction result according to the practical conditions. Industrial data validation shows that the intelligent fusion model can predict the copper matte grade more effectively compared to the single model and provide optimal control of the copper flash smelting process with potent guidance.
出处
《化工自动化及仪表》
CAS
2007年第4期26-29,52,共5页
Control and Instruments in Chemical Industry
基金
国家自然科学基金资助项目(60634020)
国家"973"计划项目(2002CB312200)
关键词
智能融合建模
铜闪速熔炼过程
冰铜品位
模糊神经网络
受约束梯度下降法
intelligent fusion model
copper flash smelting process
copper matte grade
fuzzy neural network
constrained gradient descent algorithm