机构地区: 中南大学信息科学与工程学院
出 处: 《冶金自动化》 2004年第2期44-47,共4页
摘 要: 液态金属温度高、腐蚀性强 ,流量不能直接测量和控制 ,本文提出了用神经网络建立液态金属重量与流量的数学模型 ,对液态金属流量进行软测量的检测方法 ,再用两级计算机控制系统 ,实现液态金属流量的稳定控制。仿真和实际应用结果表明 ,用该方法检测高温液态金属流量的相对误差小于 1.5 % ,满足实际生产要求。 The flow of liquid metal can not be measured directly and controlled for its high temperature and causticity.A soft-measuring approach to measure the flow of liquid metal by constructing mathematical model of weight and flow based on neural network is presented.Two-level computer system is applied to realize stable control of liquid metal flow. Simulation results and industrial application demonstrate that the relative error of measured high temperature liquid metal flow by this method is lower than 1.5 %, meeting the need of industrial process.