The prediction of critical deposition velocity in slurry pipeline based on improved SFLA-LSSVM

The critical deposition velocity of slurry pipeline is affected by the nature of the material being conveyed,physical properties of the slurry and other factors,the calculation is complicated.So it is difficult to meet the practical requirements in terms of application and accuracy by using the existing various types of empirical formulas.The present paper is aimed at combining an improved Shuffled Frog Leaping Algorithm (SFLA) with Least Squares Support Vector Machine algorithm as a new method to predict the critical deposition velocity.For the algorithm is easy to fall into premature convergence and local optimal problem,the improved shuffled frog leaping algorithm introduces nonlinear decreasing inertia weight by using particle swarm optimization algorithm in the update strategy.In addition,the algorithm combines the local search update strategy based on the average thought and the determination of local optimum of the fitness variance based on the group.The simulation results show that the prediction result of this method is superior to the conventional methods.At the same time,it has a higher precision when compared with the selected critical deposition velocity formulas.