基于彩色图像处理的浮选尾煤灰分软测量研究

Study on soft sensing of ash content in flotation tail coal based on color image processing

浮选尾煤灰分是浮选产品的一个重要指标。针对选煤厂浮选尾煤灰分多采用离线检测而无法实现在线准确测量,以及当前浮选软测量多采用单一的灰度图像从而导致软测量模型精度及适应性较差的问题,提出了一种基于彩色图像处理的浮选尾煤软测量方法,建立了基于最小二乘支持向量机(LSSVM)的浮选尾煤灰分软测量模型。模型以不同颜色空间的彩色特征、灰度均值以及浓度特征为输入变量,以尾煤灰分作为输出变量,采用粒子群优化算法对LSSVM模型参数进行优化。结果表明:所建立的尾煤灰分软测量模型可以较好地实现浮选尾矿灰分的在线预测,引入浮选尾矿图像的彩色特征可以提高尾煤图像分析的精度,预测精度达96.89%。研究成果在柳湾选煤厂现场应用,并取得了较好的尾矿灰分测量效果。

The ash content of flotation tailings is an important index of flotation products.In view of the problem that the ash content of flotation tailings in coal preparation plant is usually measured off-line and can’t be accurately measured on-line,and the current flotation soft measurement is usually based on gray-scale image only,which leads to poor accuracy and adaptability of the soft measurement model.A soft-sensing method of flotation tailings is proposed based on color image processing method,and a soft-sensing model of flotation tailings ash is established based on least squares support vector machine(LSSVM).The model takes the color features of different color spaces,gray mean and concentration characteristics as input variables,and the ash content of tailings as output variable.Particle swarm optimization(PSO)algorithm is used to optimize the parameters of LSSVM model.The results show that the soft-sensing model of tailings ash can predict the ash content of flotation tailings on-line.The accuracy of tailings image analysis can be improved by introducing color features of flotation tailings images,and the prediction accuracy reaches up to 96.89%.The research method was applied in Liuwan Coal Preparation Plant,where the ash content of flotation tailings was accurately measured.