Location:Home > Engineering science > Metallurgical Engineering > Non-ferrous metallurgy > An Integrated Optimal Control System for Aluminum Hydroxide Gas Suspension Calcinations

An Integrated Optimal Control System for Aluminum Hydroxide Gas Suspension Calcinations

Downloads: []
Tutor: Li DingFengQi
School: Central South University
Course: Non-ferrous metallurgy
Keywords: Gas suspension calcinations,Numerical Simulation,Process Control,Neural Networks
CLC: TF821
Type: PhD thesis
Year:  2008
Facebook Google+ Email Gmail Evernote LinkedIn Twitter Addthis

not access Image Error Other errors

At present, China's aluminum industry is developing rapidly, alumina production has reached 19 million tons / year. Around energy conservation, to carry out the technical innovation of the alumina industry demand is increasingly urgent. Alumina calcined alumina production, the quality and the production of energy have a significant impact on the process, the process has been widely used gas suspension calcination process. The many gas suspension calcination production shows that the process is still a great potential for improvement in the areas of device configuration, operation regulation and process control. The roasting process to carry out the optimization of the equipment, operation and control is conducive to increase production, energy saving and consumption reduction roasting production. Funded by the National Natural Science Foundation of China, with an annual output of 50,000 tons of gas suspension calciner test object, the integrated application of FLUENT, artificial neural networks, genetic optimization, fuzzy control, expert system technology, carried out on alumina calcination process equipment, control and direction of the overall optimization. Research results are: (1) roasting combustion system the missing configuration basis to carry out the study of the furnace combustion optimization simulation. On the main furnace P04 simulation study using FLUENT: an optimum air-fuel ratio of the fuel value (A / F), and hypoxia complete combustion of the corresponding optimum operating conditions; Best unloading area ¢ô Ministry furnace, V08 the preheat burner arrangement area ¢ò Ministry furnace; maintain V08 small proportion invested burner fuel advantageous to reduce NO production; improve air preheating temperature energy-saving effect. Simulated NO x , CO, CO 2 emission generation, providing an important reference for the production operation. (2) lack of analysis for roasting cyclone conditions, to carry out the study of gas-solid separation. Preheat cyclone P01 Reynolds stress transport model for solving the gas field, the Lagrange solving particle trajectories. Calculate the different conditions the separation efficiency of the cyclone, wind speed, temperature, air leakage rate and the physical structure of the P01 circulation cyclone dust collector lock gas equipment modification program provides optimized reference operation. (3) lack of existing description roasting process model proposed using neural network (ANN), genetic algorithm (GA) and gray model (GM) optimization modeling to establish the temperature forecast, exhaust soft measurement evaluation and capacity assessment of the three process model. The temperature prediction model GM (1,1) and ANN portfolio optimization to achieve absolute error is ¡À 5 ¡æ evaluation model prediction hit rate of 90% or more, can guide the production adjustment. Exhaust soft measurement model structure for ANN {3-5-4}, with the absolute error is less than 1 evaluation model prediction accuracy rate of 88.6%. Based on the the waste discharge predicted by FLUENT simulation results of the new conditions, with secondary simulation. Capacity assessment model structure ANN {3-9-1}, with a relative error of less than 1% of the evaluation model, the prediction accuracy rate of 96%. Capacity ANN model is better than the regression model reveals systematic relationship. (4) inadequate for the roasting process conventional, single PID control mode, the the established roasting process fuzzy expert control system. Design a Complex-PID controller and the air-fuel ratio Experts regulator and a roasting process segmented regulation and control strategies. Wherein the controller unit by FNN PID unit and adjust the threshold units, using fuzzy method, neural networks and genetic algorithms PID adjustments to ensure optimal or suboptimal control parameters. Optimization feedback regulator numerical simulation, video surveillance, and flue gas oxygen regulator. Segmented regulation and control policy under different conditions optimization of temperature control accuracy of ¡À 5 ¡æ, stable furnace conditions. (5) imperfect roasting production and management, and architecture of the roasting process the ANNES guidance system. Production rules, said the process of explicit knowledge, ANN model represents the implicit knowledge, two types of knowledge by the membership function to achieve transformation. Create a fan failure, the burning regulation and state analysis Knowledge Base, analysis and monitoring of the combustion process; build GA-ANNES optimization model library, process energy consumption analysis, to solve the high-yield and low consumption parameter optimization problem; build cyclone operation guidance Knowledge Base, cyclone ANNES analysis and diagnosis and operation optimization. (6) development of PLC-based SCADA systems and integrated optimization system based on the VC, Matlab. OPC technology is used for communication between the two systems, custom Agreement and DeviceNet bus achieved. PLC system-based control and optimization system integrated neural networks, genetic algorithms, expert systems for process optimization and control. Integrated optimization system developed in this paper in the annual production capacity of 50,000 tons of gas suspension calciner industrial trials to achieve good optimization effect: heat consumption decreased by 14.3% to reach a 3.09MJ/kg 8.8%; reduce the temperature of the main furnace control in 1040 ¡À 5 ¡æ; oxygen content reduced 75% in 1 to 2% of 53.9%; NO emission reducing, control of 53 ppm.
Related Dissertations
Last updated
Sponsored Links
Home |About Us| Contact Us| Feedback| Privacy | copyright | Back to top