| The distribution of coal particle size is directly related to the economic and safe operation of coal-fired boilers,which is an important parameter for stable combustion of coal-fired boilers and an important basis for the adjustment of milling system.Now,the main method to measure coal particle size is the offline sampling and analysis measurement,which is used by most of the power plant.This method not only can not accurately determine the fineness of coal in real time but also can not be adjusted boiler and milling system.Therefore,the on-line measurement of coal particle size distribution has very important value for industrial application.The measurement of coal particle size distribution online by image method has some advantage of simple,intuitive,high precision and strong anti-interference ability,etc.However,due to the technical difficulties,such as high flow rate,high concentration,and pollution of optical components,about the measurement of pulverized coal in industrial pipeline,image method has not been widely used in industry.In this paper,an on-line measurement method of automatic sampling coal particle size distribution based on image processing is designed and researched from three aspects:the construction of experimental system,the research of image processing algorithm and the on-line measurement of pulverized coal particle size.Firstly,the automatic sampling pulverized coal particle size measurement system is designed according to the principle of constant velocity sampling and boundary wind protection.Secondly,based on the image processing algorithm of Otsu automatic threshold,an image processing software with high speed identification of pulverized coal particles is developed.Finally,through the on-line measurement experiment of pulverized coal particle size,it is proved that the method proposed in this paper is correct and feasible,and has good anti-fouling ability,which is very important for the industrial application of on-line measurement of pulverized coal particle size distribution by image method. |