Abstract:

To improve the prediction accuracy on consumed electricity of manufacturing in beer enterprise, we design a method on constructing prediction model based on support vector machine and PSO(Particle Swarm Optimization) algorithm. The radial basis function is used as the kernel function in SVM(Support Vector Machine）. The paper uses K-fold cross validation and optimizes the penalty parameter c and the parameter g based on PSO(Particle Swarm Optimization). Based on the training set, which is consist of consumed water in the process of manufacturing and consumed electricity in the process of manufacturing for 28 days, and the prediction set, which is consist of consumed water in the process of manufacturing for 10 days, the paper respectively constructs the SVM prediction model on consumed electricity in the process of manufacturing based on radial basis function and polynomial function to predict the consumed electricity in the process data. The test results show that the prediction accuracy of the Support Vector Machine prediction model is 51.495% based on radial basis function kernel higher than that based on polynomial kernel in predicting consumed electricity in the process and the method is practical.

Key words: support vector machine(SVM), prediction model of consumed electricity in the process of manufacturing, kernel function, particle swarm optimization(PSO)