Abstract
In order to improve the efficiency and accuracy of corrosion measurement of circulating cooling water in petrochemical enterprises, Reduce measurement errors caused by human measurement and save measurement costs, this article analyzes the three-year measurement data of a petrochemical company. Since the Generalized Regression Neural Network (GRNN) still has a good fitting effect on prediction of noisy data, therefore, it was chosen as the main body of the forecasting model. Principal component analysis (PCA) is used to reduce the dimension of GRNN input sample parameter information, extract principal components, initially build the PCA-GRNN algorithm model, and using chaotic immune particle swarm optimization (CIPSO) algorithm to optimize the smoothing factor in GRNN, Corrosion prediction of actual production data of circulating cooling water. Experiments show that compared with the PSO-GRNN algorithm, this optimization algorithm has higher prediction accuracy and generalization ability.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call