Modified PSO Algorithm And Its Application In PTA Solvent Dehydration Tower

Particle Swarm Optimization comes into being with the research of the campaign group between birds and fish. It is a kind of adaptive random algorithm based on group search strategy, and is a new branch in the field of evolutionary computation. Its main feature is simple, fast, not many parameters need adjustment, and no gradient information that demonstrate great potential in engineering practice. This method has been widely used for function optimization, neural networks, fuzzy control, pattern recognition and other fields.Based on the basic principle, algorithm process and parameter setting of PSO, the paper has done a systematic exposition and discussed two standard algorithms: inertia weight standard model and shrinkage factor model. PSO standards are all focused on how to make PSO effectively to search the optimal solution in the solution space. However, in the high-dimensional complex optimization problems, premature convergence and convergence of the poor accuracy are still existed.Because the PSO is easy to fall into a local optimal solution, an improved PSO is proposed on the basis of the previous studies. Through changes in variation of particles dynamicly, the ability of the overall search algorithm is enhanced. At the same time, simulation experiments showed that the advantage in the convergence of PSO over standard PSO and inertia weight linear descent algorithm LPSO.PTA solvent dehydration tower is the key PTA production equipment, a smooth, safe and high-efficiency operation plays a decisive role in the production process that the solvent dehydration tower optimization is of great theoretical and practical significance. The paper studies solvent dehydration tower PTA plant, adopts PCA (principal component analysis) using a combination of BP neural network to establish soft measurement model of the acetate yield at the bottom of the solvent dehydration tower. On the basis of the method, the improved PSO is used to optimize the operation of the solvent dehydration tower. The specified acetate yield is the objective function, the optimal operation is the proposition, that are the way to acquire the optimal operating conditions for the solvent dehydration tower. Applications show that better access for the acetate yield is acquired at the bottom of the solvent dehydration tower compared with the original operating conditions.