Abstract: A computation model of fractional-order differentiators based on functional network and the learning Algorithm of functional network using particle swarm optimization was proposed. it provided a new method to compete the numerical calculation of fractional-order differentiators in other fields. the simulation results demonstrate that the method presented in the paper is more efficient and feasible. [Copyright &y& Elsevier]
Guangjin, Peng, Weidong, Meng, Duan, Cao, Kunyao, Xu, Yunhua, Yang, and Ji, Zhao
Subjects
MACHINE learning, INTELLIGENT agents, SUPPORT vector machines, ARTIFICIAL neural networks, ALGORITHMS, SIMULATION methods & models, DATA analysis
Abstract
Abstract: Analysis of power cost data belongs to small sample problem, because it is difficult to collect such type data. Many data mining algorithm and theory such as neural network cannot acquire good analysis effect. In this paper, based on SVM, modified algorithm and its model of SVM is introduced, moreover, such algorithm and its model is used to data mining and knowledge discovery of power cost data. Simulation results in MATLAB show that such algorithm and its model have high prediction exactness in power data analysis and can supply scientific and reasonable reference to control power cost. [Copyright &y& Elsevier]