CAJ | 학술논문

现场采集的实际回转窑窑头图像中熟料填充率直接反映了回转窑内料层的厚薄和传输情况,测量熟料填充率的大小和变化可用于判定窑内工况如给料量大小、烧结带温度异常波动.为了快速有效分割出熟料区,采用自适应窗口叠加空间信息生成线性权重和图像,消除了原图像的噪声同时有效地保留了边缘信息,将传统的FCM算法基于像素个数的聚类转化为基于灰度级聚类,降低计算的复杂度,利用灰度连通性采用区域生长的方法恢复熟料区,并实现熟料填充率的实时测量.最后通过对工况实例的分析来证明此方法的实用性.
현장채집적실제회전요요두도상중숙료전충솔직접반영료회전요내료층적후박화전수정황,측량숙료전충솔적대소화변화가용우판정요내공황여급료량대소、소결대온도이상파동.위료쾌속유효분할출숙료구,채용자괄응창구첩가공간신식생성선성권중화도상,소제료원도상적조성동시유효지보류료변연신식,장전통적FCM산법기우상소개수적취류전화위기우회도급취류,강저계산적복잡도,이용회도련통성채용구역생장적방법회복숙료구,병실현숙료전충솔적실시측량.최후통과대공황실례적분석래증명차방법적실용성.
The filling percentage of clinker in the real image recorded from the discharge end of rotary kiln directly indicates the thickness and transportation of material in rotary kiln. The measurement result of filling percentage of clinker can be used to reflect the real-time industrial conditions such as the feeding rate of raw slurry and the muta-tion of sintering zone temperature. In order to fast and accurately extract the region of clinker from the image, this paper develops an innovative FCM method that applies local spatial neighborhood information with adaptive win-dow to generate the linear-weighted sum image, in which the noise is eliminated while the edge is preserved. It is worth to note that the clustering segmentation used in this paper is based on the number of gray-level rather than the number of pixels of the image to reduce the computational complexity. The total clinker region is then recognized by region growing technology based on gray-level connectivity. Finally, real-time measurement of filling percentage of clinker is completed by calculating the rate of number of pixels between clinker region and the total discharge end of rotary kiln. Case study demonstrates the applicability of the proposed approach.