Browsing by Author "Boadu, Kwaku Debra"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Adaptive control of alumina concentration in the hall-heroult cell using neural network
    (1996) Boadu, Kwaku Debra
    Aluminium smelting the world over has had two major constraints in recent years: environmental protection and energy costs. Since the method and efficiency of alumina feed in the smelting process impacts environmental pollution and production efficiency greatly, much of the industry’s investment money has been spent in researching into better feed control systems - feed delivery systems and feed strategies. The subject matter of this thesis dwells on the latter, and continues the search for an efficient adaptive alumina feed strategy in the Hall-Héroult cell for the reduction of aluminium. Neurocomputing, one of the fastest growing control system theories in contemporary electrical engineering, is applied to the problem of on-line estimation of alumina mass balance in the electrolytic cell. A contribution is proposed to alumina feed control strategies by developing a neural network-based adaptive feed control algorithm, robust against cell resistance variations, and implementable on retrofit state-of-the-art aluminium reduction cell microcomputers. Electrolytic resistance/alumina concentration data from a simulated l4OkA Center-Break cell was used as input vectors to train a single-layer feed forward loop-back NEURAL NETWORK1 constructed with six constraint equations and six degrees of freedom. The identified prediction algorithm was successfully tested on both simulated and real cell resistance data and results presented. The algorithm was also compared with the extended Kalman filter using the same test bed and shown to be a better solution to the problem under research. Finally a NEURAL network-based feed control strategy (NetFeed) is developed and presented in this thesis.