【摘要】 利用模糊神经网络建立了具有时间序列对象的预测模型，提出了相应的模糊化方法，并对我国农机总动力需求进行了预测，预测结果和实际情况有较好的一致性更多还原

【Abstract】 Based on the combined merits of fuzzy inference mechanism and neural network, this paper presents a fuzzy neural network (FNN) model for forecasting time series problems. With rapid development of techniques in neural network and fuzzy logic systems, the fuzzy neural networks are attracting more and more interests since it is more efficient and powerful than either neural networks or fuzzy logic systems. FNN can accelerate the convergence speed of the system; it can also access to any process adaptively with any accuracy without an analysis and synthesis of the mathematical model of the process. This paper proposes a general method to generate the fuzzy membership functions from numerical data. In order to simplify the algorithm and realize fuzzification, the input and sample data are processed by a special method. Obtained results show that the forecasting values from the model agree well with the original values.更多还原