Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Automation systems > Automatic control,automatic control system

Design and Research of Optimal Control System for Cement Pre-grinding Process

Author LiuZhiPeng
Tutor YanWenJun
School Zhejiang University
Course Control Theory and Control Engineering
Keywords Pre-grinding system FLC LSSVM LMI MPC DCS C++
CLC TP273
Type Master's thesis
Year 2012
Downloads 80
Quotes 0
Download Dissertation

Pre-grinding system is one of the key components in the operation of cement plants which significantly promotes the energy utilization efficiency. Due to its technical advances and economic feasibility, it has been widely adopted in industry and brought about tremendous benefits. At present, control solutions dedicatedly designed for pre-grinding systems are limited in literature, moreover, they are mainly carried out in terms of theoretical models or simulation assessments without being adopted and validated in industrial fields..Pre-grinding system needs to be stably controlled for operational efficiency improvement and energy saving in cement plants.Pre-grinding system is a complex system with a series of features, such as strong coupling, nonlinear, multivariable and large delay. As a result, it is difficulty to control using only one algorithm. In order to deal with this problem, an intelligent optimal control system is built to stabilize and optimize the pre-grinding system in this paper. The research of this paper is based on the subject of state 863 project:energy saving and optimization of typical equipments in cement industry. The main research contents of this paper are as follows:1. Based on the analysis of pre-grinding process and specifications, the key parameters and the main control scheme are defined.2. According to the pre-grinding operational experiences, Fuzzy Logic Control (FLC) is used to make the bin weight stable. An intelligent fuzzy logic controller is designed to control the bin weight, including defining the input (error of bin weight and error change of bin weight), output (feed rate) and the membership function of variables, data preprocessing and building fuzzy rules.3. According to the control demand of pre-grinding system, the model is obtained by Least Square Support Vector Machine (LSSVM) regression, and Linear Matrix Inequality Based Model Predictive Control (LMI-Based MPC) is used to control the system load. A model predictive controller is designed to optimize the system load, including building the model between input (classifier current) and output (fan speed) by LSSVM, obtaining the control law of model predictive controller by LMI.4. The control system software is developed by C++ and relevant technologies. It can be applied to the industry filed with the support of DCS and OPC.5. The control system simulation has been performed using the data collected from the field. And the result demonstrates that the pre-grinding system is more stable with the control system.The design process of control system is innovative in this paper. It is of great meaningful to the cement plants for energy saving and emission reduction.

Related Dissertations
More Dissertations