论文部分内容阅读
针对目前转炉平台氮封控制系统的缺陷,运用CCD辐射图像处理技术,通过采集某炼钢厂22.5 m炼钢转炉平台氧枪口不同喷溅强度的火焰图像,建立了Kohonen自组织特征映射网络智能聚类氮封控制模型。工业应用表明,该模型正确、可靠,实现了炼钢转炉平台氧枪口氮封实时控制。
Aiming at the defects of the nitrogen sealing control system of the converter platform at present, the image of Kohonen self-organizing feature mapping network is established by using the CCD radiation image processing technology. By collecting the flame images of different sputtering intensity at oxygen muzzle of 22.5 m steelmaking converter platform in a steelmaking plant, Clustering nitrogen seal control model The industrial application shows that the model is correct and reliable, and the real-time control of the oxygen muzzle nitrogen seal in the steelmaking converter platform is realized.