Intelligent Optimization Of Operational Parameter In Nitric Acid Process

For device of chemical process, because many factor such as the price of the material, production environment, requirement of products and, performance of device are varying and impossible to keep the same as specified in original design, in these cases, it is necessary to adjust produce plan and operational variables so that it really works in an optimized state to cope with the changing condition. The conventional steady-state optimization method is on the relied precise mathematical model, many processes is presented non-linear in very degree, and it is quite complex in internal mechanism; in addition along with the complex chemical process, it is very difficult to get the precise mathematical model. The paper is paid to explore the modeling and optimization methods which need not know the process accurately, and only need the input and output data, it can improve efficiency of the yielding, reduce yielding cost and increase economical benefit.In this work, we will emphasize our study up on the problem of optimization technique, and will not change primary equipment and disturb natural yield, the better use of existing productivity will be explored, and finally to raise the quality of products and decrease the operational cost of the plant. The paper is generalized necessity of optimization and combine scientific research and enormous benefit in chemical process, and is simply discussed operational optimization. Take example for nitric acid of a plant, based on the neural network, the modeling for the steady-state process are studied and genetic algorithm is discussed in this dissertation, the research work of this thesis consists of the following major parts:(1) According to the current operation statues of nitric acid of a plant, the major controllable independent variables, dependent variables and the object function of optimization problem was formed.(2) Mathematical model is regarding one of the core work in operation optimization using the neural network technology, the neural network provided with a lot of virtue, such as high rude, association of ideas and high non-linear mapping, described complexity of Chemicalprocess. Therefore, a high quality identification system model using forward feedback network BP algorithm is obtained by combining the actual spot data.(3) Genetic algorithm is essentially a kind intelligent searching optimization technology. With the precondition that the neural network model can accurately reflect the operation situation of the systems, the paper discusses how to effectively integrate the neural network and genetic algorithm, by utilizing the comprehensive searching ability to maximal object function of the system under a restriction condition, the controlling state parameters of the system and the calculation results indicate the optimization result of a steady state is satisfactory.