摘要
把模糊逻辑系统与神经网络相结合,形成结构像神经网络,功能似模糊逻辑系统的模糊神经网络系统,该系统具备了模糊逻辑系统和神经网络的优点,克服了单个系统的不足。再结合误差反向传递学习算法(BP算法),调整模型参数及权值。最后应用模糊神经网络系统解决实际问题,经过若干次学习训练,使系统达到稳定,通过仿真结果可看出,将所设计的模糊神经网络系统应用在WTI原油价格预测中具有可行性与有效性。
A kind of fuzzy neural network system which combines fuzzy logic system and neural network was designed. Its structure as neural network and its performance like fuzzy logic system is very like each other. The system has the advantages of both the fuzzy logic system and the neural network that overcome the shortages of the single system. Then using the error reverse transfer learning algorithm (Back-Propagation algorithm) adjusted the system parameters and weights. Finally, the fuzzy neural network system was used to have solved the actual problerrL The system went through many times training and made the system stable. The simulation results can show the design of the fuzzy neural network system has feasibility and effectiveness. If used to forcast the price of WTI Crude Oil, the system being concerned is feasible and valuable.
出处
《辽宁工业大学学报(自然科学版)》
2013年第5期347-350,共4页
Journal of Liaoning University of Technology(Natural Science Edition)
基金
辽宁省百千万人才资助项目(2012921055)
关键词
模糊逻辑系统
神经网络
BP算法
fuzzy logic system
neural network
Back- Propagation algorithm