Modeling and control of ph in industrial sodium chlorate cell using soft computing techniques

Abstract

The growth of the sodium chlorate industry is promising since it is globally used to manufacture newlinechlorine-based bleaching agents, perchlorates used as rocket oxidizers, and in newlineagricultural and milling applications. Sodium chlorate production is one of the most extensive newlineenergy-intensive industrial-scale electrochemical processes, where power consumption newlineaccounts for over 70% of the production costs. Hence, there have been constant newlineefforts to improve sodium chlorate production in terms of quantity and quality and thus newlineimprove energy efficiency. newlineSodium chlorate is manufactured by electrolysis of hot acidulated brine. The kinetics newlineof this reaction depends on the pH and local concentration of active chlorine. The reaction newlineis not immediate, and if pH is not controlled, secondary reactions occur in the cell, leading newlineto a significant loss in the current efficiency. Hence it is essential to model and control newlinethe pH of the electrolytic cell to improve the quantity and quality of the product as well newlineas power efficiency. This research proposes modeling and control of the pH process in newlinethe chlorate cell. This is promising as most of the research related to this system reported newlinemethods to improve the process using different cell designs, selection of anode and cathodes, newlinebuffers, etc. So far, no studies have been reported on predicting and controlling the newlinepH of the chlorate cell and thus improving power efficiency. In this regard, the present newlineresearch focuses on sodium chlorate cells with two-fold goals: to develop a model of newlineindustrial-scale chlorate cells and design a controller to regulate the pH of the cell bulk. newlineArtificial intelligence methods are superior to traditional methods for modeling, especially newlinein chemical processes, as it is very complex due to the presence of many nonlinear newlinevariables such as pH and dead time, uncertainty in the parameters, and the nonlinear relationship newlinebetween various parameters. newline