论文部分内容阅读
为解决赤铁矿强磁选过程运行因工艺生产指标无法在线连续检测而难以实施闭环优化控制的问题,将过程知识、专家经验与过程数据相结合,设计了由基于案例推理的回路预设定模型、小波神经网络软测量模型以及专家规则补偿器构成的智能优化设定控制器.针对已有优化控制软件存在的标准化程度低、可扩展性差、资源无法复用等问题,设计并开发了基于.NET组件技术的集优化控制与运行监视为一体的磁选优化设定控制软件系统.该系统针对具有不同编程技能的研究人员,实现对动态链接库、Matlab脚本以及VBScript脚本3种算法形式的支持.在半实物仿真平台的实验研究表明,所开发的系统具有一定的鲁棒性,有效保证了最终产品质量.
In order to solve the problem that it is difficult to implement closed-loop optimal control due to the inability to continuously measure the production index of hematite during the process of strong magnetic separation, the process knowledge, expert experience and process data are combined to design a loop pre-set based on case-based reasoning Model, wavelet neural network soft sensor model and expert rule compensator.Aiming at the problems of low standardization, poor scalability and unavailability of resources, the existing optimization control software has been designed and developed based on .NET component technology set optimization control and operation monitoring as one of the magnetic separation optimization setting control software system.This system aimed at researchers with different programming skills to achieve dynamic link library, Matlab script and VBScript script in three forms of algorithm The experimental research on the semi-physical simulation platform shows that the developed system has certain robustness and effectively guarantees the final product quality.