Font Size: a A A

Application Research Of Advanced Process Control Theory In Complex Industry Control System

Posted on:2007-09-08Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y YuanFull Text:PDF
GTID:1118360245483134Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
The complex industry control system has the characteristics such as numerous relate variables, complex process and uncertainty factors, serious correlative coupling, multiform process information and etc. So, it is difficult to modeling and control the complex industry control system by traditional control methods, which makes the complex industry control system become the very important research object of control theory and engineering region. The requirements of high control performance of enterprises and fast development of control theory and computer technology have boosted the in-depth researches and applications of advanced process control theory.The pre-baked aluminium electrolysis produce process is selected to be the main research object of this dissertation as a typical complex industry control system, which is a nonlinear, multivariable coupling, time-varying and big time-delay industry process system. Many parameters and variables of the aluminium electrolysis process have fuzzy uncertainty and cannot be measured continuously because of the complex physical and chemical changes and the disturbance of the environment condition and manual works. It is difficult to realize optimal control of the aluminium electrolysis process by adopting the traditional control methods based on mathematical model, so much as the single intelligent control method.Considering the shortcoming of current systems, this dissertation has deeply studied the intelligent control technology, such as fuzzy control, expert system, neural network, support vector machine, chaos theory and etc. The design mode of adopting parameters predictive model and integrated control technology is proposed, and based on detail analysis of mutuality theory basis and design methods, the pre-baked aluminium electrolysis produce process networked control system is designed. The research production has successfully applied in industry field.The main research contents and achievements of this dissertation are listed as follows. (1) Considering pre-baked aluminium electrolysis cell has the problems of nonlinear, multivariable coupling, time-varying, big time-delay and has difficulty to modeling accurately, the neural network modeling method based on chaos optimal is adopted. This method can improve accuracy and generalization ability of neural network. The important parameters prediction models are constructed, such as alumina concentration, electrolyte temperature and anode distance, which has very important significance to aluminium electrolysis process control.(2) Based on the analysis of the advantages and disadvantages of fuzzy control in theory, a new adaptive fuzzy algorithm, which integrates chaos optimal and fuzzy control, is proposed. Controller parameters, such as the quantifying factors and the scale factor, and fuzzy control rules are optimized by chaos method, which was a combination of global and partial chaos searching. The algorithm convergence is proved, and the stability of fuzzy control system is analyzed. Simulation research is done by MATLAB, and the results proved the efficiency of this algorithm.(3) Support vector machine is applied in the expert system of aluminium electrolysis cell status resolution and diagnosis. The expert system structure based on multi-SVMs decision-making model is proposed, and the basis principle is discussed. The expert system has abilities of cell status resolution, cell status diagnosis, control decision-making and parameters optimizing automatically.(4) The collectivity structure of aluminium electrolysis process advanced control system is designed, by use of the control ideal of advanced process control and integrated control. Based on collecting control experience and knowledge of experts, the NN-PID control strategy of anode distance, the fuzzy control strategy based on chaos optimal of electrolyte temperature and the expert fuzzy control strategy of alumina concentration are proposed.(5) Considering the optimal control of important parameters, output, quality and energy consumption, the intelligent coordination strategy based on fuzzy-expert control technology is proposed, which can satisfy global optimal control target.(6) The networked control system is studied from two aspects, which is theory analysis and technology realization. The definition of networked control system is introduced, and its characteristics and advantages are discussed. Then, the pre-baked aluminium electrolysis process NCS is designed. The influence of system's stability by time-delay and data packet dropout is analyzed, and the stability condition is studied. At last, the system's topology, hardware and software are designed.Finally, based on the summarization of research works in this dissertation, the further development is proposed. What's more, the future is expected.
Keywords/Search Tags:complex industry control system, advanced process control (APC), chaos neural network, expert fuzzy control, support vector machine (SVM), networked control system (NCS)
PDF Full Text Request
Related items