Research Of Coal Load Control Method Based On Ball Kinetic Energy For Ball Mills

In the energy-based sustainable development society, the thermal power is key to reduce energy consumption of industrial production. The boiler as the major coal combustion equipment of the thermal power plays a decisive role. The ball mill is the key equipment in the boiler combustion control system. A better solution for the detection and control of the coal load is the priority in the optimization of the coal pulverizing system. There has not been a unified, rigorous and complete mathematical theory for a ball mill grinding process, and this theory lacks a more complete and accurate method for supervising and controlling the coal load. Therefore, it is an urgent task to improve the automation level of a ball mill coal pulverizing system.To date, no research has addressed the use of a grinding medium to reflect the coal load which is the controlled parameter of a coal pulverizing system. In order to improve the performance of the coal load detection in a coal pulverizing system of power plant, a novel coal load control method based on the balls’ kinetic energy is proposed in this paper. The theories and methods of the operating efficiency of a coal pulverizing system, the coal load and the balls’ kinetic energy are obtained, which combines the operating characteristics of a ball mill and the ball motion under the influence of the coal load are analyzed carefully. It is worth focusing on the optimum control of the coal load.The theoretical research results and innovative points are summarized as follows:1. It is proposed that the balls’ kinetic energy is used to reflect the coal load which is the controlled parameter of the coal pulverizing system.(1) Correlation model of coal load and grinding efficiency. Based on the realtime operation data in power plants, the relationship between the loads that reflect the working efficiency, the coal load and the boiler operation parameters is analyzed carefully by using the correlation degree and balance degree method from the viewpoint of the basic parameter and the kinetic energy. The motion states of balls under the optimum operation conditions of a ball mill are carried out by Matlab software platform. The variation curve of the balls in projectile motion is established, and the position coordinates are determined to calculate the balls’ kinetic energy. The correlation degree between the coal load and the ball motion is obtained. The balls’ kinetic energy is more associated with the coal load of a ball mill than with the boiler operation parameters by a polynomial fitting and information fusion model. The approach provides the study with the theoretical basis for a novel optimum control method of the coal load based on the balls’ kinetic energy.(2) Energy calculation model for ball kinematics. The operating conditions and process parameters are analyzed carefully in a ball mill. With the increase of the coal load, spacial distribution states of balls are described using multizone methods. A novel energy calculation model for ball kinematics based on the coal load of a ball mill is proposed in this paper. Meanwhile, a practical experiment is carried out to estimate the validity of the model. It is confirmed that the projectile motion is the optimum state for balls to obtain the maximum kinetic energy, and there is the corresponding optimum coal load at the moment. The spacial distribution of balls is easy to be measured, consequently, the real-time kinetic energy of balls is obtained by the energy calculation model to adjust the coal load of a ball mill accordingly. It is further indicated that the balls’ kinetic energy directly reflects the coal load. The grinding process of a ball mill is studied. The approach is utilized as the theoretical and experimental basis to realize a coal load control method based on the balls’ kinetic energy.2. It is obtained that the functional relations between the operating efficiency of the coal pulverizing system, the coal load and the kinetic energy of balls are close.(1) Discrete Element Method (DEM) simulation experiments. The motion states of balls and coal as particles are analyzed by DEM based on the molecular dynamic method. Moreover, the mill operation process is simulated by PFC3D software platform based on DEM. The relationship between different coal quality, coal load, coal size distribution and the ball motion is analyzed carefully under certain quantitative optimized operating parameters configuration. Consequently, the variation law of parameters which represent the ball mill’s working efficiency is drawn by compared experiments. The parameters include the balls’real-time kinetic energy, the friction energy consumption, and the mill’s total work. The results indicate that the impact energy balls obtained in the projectile motion is more than in the cascading motion, and the use ratio of the balls’ kinetic energy is first increasing and then decreasing as the coal load increases. A close relationship between the coal load of the ball mill and ball movement is found.(2) Ball mill physical model experimens. The ball motion under the influence of the coal load is fully analysed by using Graphical User Interface (GUI) image edge detection method and the tool of Origin. With the increasing coal loads, balls and coal loads in the mill are going through the cascading motion, the projectile motion of several balls and the one of most balls. It is indicated that the more impact energy balls obtain in the projectile motion state, the higher the grinding efficiency is.The validity of the coal load control method based on the balls’ kinetic energy is estimated by practical experiments.3. The coal load control strategy based on the balls’ kinetic energy is proposed in this paper.(1) Prediction with balls’ kinetic energy for coal load. Based on the PFC3D simulation experimental data of the ball motion and the coal load of a ball mill, the neural network model of information fusion for the coal load and the kinetic energy of a ball mill is established by using an adaptive combination learning algorithm. The coal load in the mill cylinder is predicted from the energy point of view. The neural network information fusion model has good predictive power to the coal load. This method provides the effective experimental basis for a novel optimum control method of the coal load based on the kinetic energy of balls. Moreover, the system parameter model for the control of the coal load based on the balls’ kinetic energy is established. The step response experiment using the data of balls’ kinetic energy under step disturbance of coal loads is carried out, and the transfer function model of a ball mill coal pulverizing system is conducted on the basis of dynamic system identification. A coal load prediction model based on the balls’ kinetic energy is estimated, and the operating characteristic of a coal pulverizing system is described by the recursive least-squares method (RLC). The results show the high prediction precision and accuracy rate of RLC for the coal load.(2) The correlation model of the kinetic energy of balls and coal load for ball mills is established by conducting the balls’ real-time kinetic energy. The coal load control method based on the balls’ kinetic energy is proposed in this paper. This approach decreases the difficulty with the direct detection of the coal load. In comparison to these existing coal load control methods, a novel method based on balls’ kinetic energy decreases the influence of other environmental factors such as coal quality, air flow and background noise on the coal detection and enhances the accuracy of coal load control.