论文部分内容阅读
运用数字图像处理技术 ,对烧结机尾的断面图像进行了处理和分析 ,并结合工艺研究成果、专家经验和常规仪表数据 ,解决烧结矿产量的在线推断问题 .安阳钢铁公司烧结厂的现场实验结果表明 ,通过适当的图像模糊增强 ,采用Marr -Heldreth算法着重对烧成带、燃烧带和生料层进行边缘检测等处理 ,系统达到了接近于 1级的预测精度 .该方法还可用于解决烧结终点的闭环控制 ,以及与纹理和颜色相关的质量指标的预报 .
Using the digital image processing technology, the cross-sectional images of the sintering machine tail are processed and analyzed, and the on-line inference of the sinter output is solved by combining the research results of the technology, the expert’s experience and the conventional instrument data.The field experiment results of the sinter plant in Anyang Iron & Steel Co., The results show that the system achieves a prediction accuracy close to level 1 by using the Marr-Heldreth algorithm focusing on the edge detection of the sintering zone, combustion zone and blanket layer with appropriate image enhancement. This method can also be used to solve the problem of sintering Closed-loop control of end points, and prediction of quality indicators related to texture and color.