基于图像处理的皮带煤量动态计量技术及在皮带调速系统中的应用研究

发布时间：2018-06-27 12:07

本文选题：皮带调速 + 图像处理　；参考：《山东科技大学》2017年硕士论文

【摘要】：我国是世界煤炭生产大国,煤炭的皮带输送是煤炭开采中一个重要环节,皮带输送机的功耗占整个矿井容量的30%,依据皮带上的煤量对皮带输送机进行实时变频调速来实现煤矿节能降耗生产,成为一种发展趋势。针对目前采用皮带秤实现皮带上煤量动态计量的方式存在造价高、安装维护工作量大等问题,本文设计了一种基于图像处理的皮带煤量动态计量方法:对上传到地面上的矿井皮带监控视频流进行实时图像处理,计算出煤宽比,依据皮带上煤量估算公式估算出煤量。该方法在无须增加硬件成本,且其计量精度能够满足变频调速要求。本文以现场的实际图像数据为研究对象,按照图像处理流程进行分析。在图像预处理阶段,针对图像反光的问题提出了一种基于窗口阈值的自适应滤波方法,有效去除噪声的同时保留了边界信息。在图像分割提取阶段,使用模板匹配法,选取反光程度不同的两个模板,利用EMD算法,把图像分为反光严重图像和反光微弱或无反光图像两类,每一类根据其特点采用不同的处理方法。针对边界对比度低和现场光源不稳定的问题,改进了自适应阈值分割算法,提高了分割效果,并采用先腐蚀后膨胀的方法对图像进行处理,提高了边界识别精度。最后,利用OpenCV图像处理库建立了基于图像处理的调速系统。该系统在兴隆庄煤矿进行现场测试,结果表明,系统的稳定性、计量精度和实时性均达到了实用的要求,为下一步通过生产应用为煤矿的节能降耗打下了坚实的基础,具有一定的推广实用价值。
[Abstract]:China is a big coal producer in the world, and the belt conveyance of coal is an important link in coal mining. The power consumption of belt conveyer accounts for 30 percent of the capacity of the whole mine. According to the coal quantity on the belt, it has become a development trend to realize the energy saving and consumption reduction production of coal mine by using the real-time frequency conversion speed regulation to realize the coal belt conveyer. In view of the problems of high cost and heavy workload of installation and maintenance in the way of dynamic measurement of coal quantity on belt by using belt weigher at present, In this paper, a dynamic measurement method of belt coal quantity based on image processing is designed. The real-time image processing of mine belt monitoring video stream is carried out, the ratio of coal to width is calculated, and the coal quantity is estimated according to the estimation formula of coal quantity on belt. This method does not need to increase the cost of hardware, and its measuring accuracy can meet the requirements of frequency conversion speed regulation. In this paper, the actual field image data as the research object, according to the image processing process. In the stage of image preprocessing, an adaptive filtering method based on window threshold is proposed to solve the problem of image reflection, which effectively removes noise and preserves boundary information. In the stage of image segmentation and extraction, the template matching method is used to select two templates with different reflective degree, and the EMD algorithm is used to divide the image into two categories: serious reflective image and weak or non-reflective image. Each class adopts different processing methods according to its characteristics. Aiming at the problems of low boundary contrast and unstable field light source, the adaptive threshold segmentation algorithm is improved, and the segmentation effect is improved. The image is processed by first corrosion and then expansion, and the accuracy of boundary recognition is improved. Finally, a speed control system based on image processing is established by using OpenCV image processing library. The system has been tested in Xinglongzhuang coal mine. The results show that the stability, measurement precision and real time of the system have reached the practical requirements, which lays a solid foundation for the energy saving and consumption reduction of the coal mine through the application of production. It has certain popularization and practical value.
【学位授予单位】：山东科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD634.1;TP391.41

【参考文献】