Study on coal logistics demand forecast based on PSO-SVR

Abstract

The coal logistics demand in this paper is refer to the demand of coal transportation, mainly including: the railway, highway and waterway freight volume of coal. In consideration of the small and the nonlinear history sample, this paper combines the support vector regression machine (support vector regression, SVR) and Particle Swarm Optimization algorithm, (Particle Swarm Optimization, PSO) to propose PSO-SVR coal logistics demand forecasting model which is suitable for the learning of small samples. Taking Coal railway freight volume for example, the paper first select influence factors and coal railway freight volumes from 1995 to 2011 as the learning samples to establish the "influence factors -coal railway freight volume" SVR model and then use the particle swarm algorithm to optimize model parameters, Finally, it forecasts the coal railway freight volume. The results show that the prediction accuracy of PSO-SVR model is superior to the BP neural network model.

Parameter optimization of SVR based on DRVB-ASCKF

“…SVM are highly successful in solving various nonlinear and nonseparable problems in machine learning [1][2]. SVM is widely used to solve the classification and regression problems [3].…”

Two Kinds of Classifications Based on Improved Gravitational Search Algorithm and Particle Swarm Optimization Algorithm

“…They have attracted the attention of more and more scholars and can effectively solve many scientific problems, such as classification and prediction. Kavoosi et al used GA to forecast global carbon dioxide emissions [8], Agrawal and Bawane used PSO to classify multispectral satellite image [16], Chen and Liu used PSO to forecast coal logistics [26], Vazquez and Garro used ABC to classify crops [27], Kang et al used FOA to classify rolling bearing fault [28], and Behrang et al used GSA to forecast oil demand in Iran [29].…”