Caustic dissolution rate ratios based prediction model parameter optimization settings ingredients Bayer Research
|School||Central South University|
|Course||Control Theory and Control Engineering|
|Keywords||Original pulp ingredients Grey Model Fuzzy Identification Liquid-solid ratio Soft Measurement|
Alumina and alumina raw slurry ingredients is high pressure digestion Bayer alumina production two important process , their mutual contact and mutual influence . Original pulp ingredients are Bayer alumina production in the first process , the task is to prepare qualified high pressure digestion process raw pulp , can prepare to meet the production requirements of the original pulp, will directly affect the dissolution rate of this caustic dissolution ratios two important economic and technical indicators . But now thanks to manual calculation operation batching parameters , and does not reflect changes in working conditions dissolution , resulting in caustic dissolution rate ratios and unstable , unable to meet the actual control requirements. Therefore, this paper proposes a ratio based on the dissolution and dissolution rate parameters caustic ingredients prediction model parameter optimization setting model , effective solution to the problem formulation process parameter optimization settings . The main research results include: ( 1 ) analysis of high alumina dissolution process on the basis of the mechanism to determine the effect of caustic dissolution rate ratios of the main factors , ratios presented caustic dissolution rate mechanism model ; then presented based on the main multi- element analysis of the neural network model, in order to establish caustic dissolution rate ratios and mechanistic models and neural network intelligent integrated predictive models. ( 2 ) in the analysis of raw slurry blending process , based on the identified impact ore slurry to solid ratio of the main factors and the relationship between them , according to the material balance principle to establish a liquid to solid ratio of mechanistic models . Mechanistic model in which the material composition parameters using gray prediction model to solve the problem of parameter detection lag . ( 3 ) In order to solve the liquid-solid ratio mechanism model does not reflect changes in working conditions during the dissolution process defects in the analysis of the mechanism of the dissolution process , based on the theory of fuzzy identification data from a large number of factory floor to dig out the pulp caustic dissolution and dissolution rate ratio ore slurry to solid ratio of fuzzy expert rules , and according to the Bayer process caustic dissolution dissolution rate ratios soft sensor model predicted values ??, mechanistic model for liquid-solid ratio correction . Simulation results show that the use of caustic dissolution rate ratios intelligent integration of predictive models for batching liquid to solid ratio parameter optimization effect is good, stable production .