Modeling of key production indices and operating parameters optimized set for sugar clarification process

Because of the random conditions and lag of assay exist in the clarification process of carbonation method sugar factory, it is often hard to obtain the optimal setting values of the critical operation parameters, which results in the key production indices difficult to meet requirements. To solve this problem, at first, a new production indices predictive model by generalized dynamic fuzzy neural network was proposed, and verified its validity through comparing with the model established by BP neural network; then find the optimal operation setting values of typical conditions by the method of multi-objective genetic algorithm on a basis of the multi-objective optimization model, and construct optimization mode library. Finally, obtain the optimal operation setting values of current condition through query and correction. The simulation results showed that, compared with the artificial setting, the way of setting operation parameters used in this paper can adapt to frequent change of conditions, and make the production indices meet requirements at any time. The optimal production can be realized.