Search

A CFD software is used to implement the numerical simulation, where two kinds of 400 kA grade aluminum cell gas collection system are studied. Based on the theory of computational fluid dynamics and energy balance, a multi-physical field model of aluminum electrolytic exhaust gas is established, the structure and the operation condition of the electrolytic cell is studied in this paper. The main conclusions are as follows: For model A, opening the tapping end side hooding panels produces larger impact on flowing exhaust gas than opening the duct end side hooding panels and keeping airtight condition, which makes the average flow rate in the two ducts decline and cause greater effect on the exhaust gas flow rate. For model B under three conditions, opening the hooding panels produces little effect, and is more conducive to maintaining gas stability of electrolytic cell.

The high temperature air combustion(HiTAC) process in gas suspension calcinations(GSC) was studied by using a CFD software FLUENT that can simulate the three-dimensional physical model of GSC with the k-epsilon turbulent viscous model, PDF non-premixed combustion species model, P1 radiation model, thermal and prompt NO pollution model. The simulation vividly describes the distributions of the temperature, velocity and consistency fields. Finally, the optimal operation conditions and igniter configuration of particular fuel combustion are obtained by analyzing and comparing the simulation results. And the emission quantity of NOx, CO and CO2 deduced from computation can play a role as reference. These optimal and estimated values are beneficial to practical operation.

The recycle fluidization roasting in alumina production was studied and a temperature forecast model was established based on wavelet neural network that had a momentum item and an adjustable learning rate. By analyzing the roasting process, coal gas flux, aluminium hydroxide feeding and oxygen content were ascertained as the main parameters for the forecast model. The order and delay time of each parameter in the model were deduced by F test method. With 400 groups of sample data (sampled with the period of 1.5 min) for its training, a wavelet neural network model was acquired that had a structure of , i.e., seven nodes in the input layer, twenty-one nodes in the hidden layer and one node in the output layer. Testing on the prediction accuracy of the model shows that as the absolute error ±5.0 °C is adopted, the single-step prediction accuracy can achieve 90% and within 6 steps the multi-step forecast result of model for temperature is receivable.