Theoretical Research And Parameters Optimization Of Vibrating Screening Based On Composite Motion Of Translation And Swing

Sieving which is one of the oldest and most extensively employed physical size separation techniques has been widely employed in many granular media industries such as mineral separation, construction, pharmaceutical and food industries. By far vibrating screen is the most common screening form. At present there are some problems in the research of vibrating screen. It is hard to design screening machine without available comprehensive model. The study of complex screening process is not thoroughly enough. A technical approach on physical tests develops slowly. To improve the existing screening theory with the purpose of increasing screening efficiency, in this dissertation the process of sieving was studied based on the 3D numerical simulation and experimental prototype. The method of using new vibrating mode(Vibrating screening of translation-swing composite motion), setting up the non-linear mapping set by Support Vector Machines(SVM) about between screening parameters, screening efficiency and processing characteristic quantities and improving the structure of the screen plate offered insights to the design of vibrating screens,these main contents as following:1). Based on the discrete element method(DEM), considering the complicate interactions of particulates and screening machine numerical simulations were established using the vibrating screen of translation-swing composite motion in this dissertation. All particles’ movement information used to calculate the screening efficiency and sieving process were recorded under different screening parameters. And the dissertation analyzed the equation of motion of screen plate and the force analysis of a single particle over the screen surface.2). The application of non-linear regression modeling based on SVM is used for mapping the sample space of parameters and vibrating screen efficiency. A comprehensive mathematical model was established to guide the design of vibrating screen. SVM successfully integrated dual representations, feature space, learning theory, optimization theory and algorithms which can solve small sample, nonlinear and local extremum problems by parameter optimization and multi-kernels. In order to meet multi-extremum, large-scale, and non-differentiable of this computational model, the intelligent algorithm with strong robustness and global convergence was applied to parameters optimization.3). Reasonable looseness distance of particles over screen plate, the size distribution of particles in the layer, collision frequency and penetration probability of fine particle, these stable screening process characteristic quantities are the main factors affecting screening efficiency. This dissertation precisely analyzed the screening process and described the screening process using four characteristic quantities: looseness, layering, touching and penetration. We established reasonable evaluation system for the four characteristic quantities, and analyzed the principle of influence vibrating parameters and structure parameters on characteristic quantities. Using the improved support vector machine, this dissertation built up mathematical models between characteristic quantities and screening parameters. A quantitative analysis of screening process details was conducted.4). Design and manufacture experimental prototype of adjustable parameters to meet the experimental requirements of vibrating screen. Experimental study about many parameters can be conducted and single motion and composite motion also can be completed. Through screening efficiency of the prototype vibrating screen under different screening parameters, the consistency between the predicted value and the experimental prototype is verified. It also verified that the innovative design is better than traditional vibrating screen which has the single motion.5). An experiment platform which combines high-speed camera system with experimental prototype of vibrating screen which has composite motion of swing and translation was designed. The research made a more in-depth investigation of particles’ movements. Because of the limitations of the large, complex and traditional calibration algorithm, the dissertation comes up with the camera calibration technique on account of neural network to obtain the relationship between two-dimensional image and three-dimensional information. Experimentally, coupled with the image processing technique, segmentation of particle motion’s image and technique of extracting image feature, the relationship between parameters with looseness distance and collision probability was analyzed. From an experimental point of view, complex physical processes were explored. When reaching a stable screening process,screening parameters were provided to improve screening efficiency. Feasible region of parameters optimization were also gained.6). The purpose is to design the geometrical structure of screen plate. DEM was used to simulate the sieving process of banana screen with the inclinations of decks improved by Fibonacci sequence. The effect of each deck along screen length on the screening efficiency and the fine ratio of particles were studied in this dissertation. Comparing vibrating screen which is widely used in industry with the experiment prototype of banana screen,the result increased by about 7% simply verified the feasibility of the methodology.