【摘要】 对电力通信网光纤保护通道风险评估进行了研究,针对电力通信网光纤保护通道风险评估的影响因素繁多,复杂性和重要程度不同,难以用精确数学模型描述的特点,提议利用主成分分析(PCA)-径向基(RBF)神经网络进行风险评估的方法。具体是利用主成分分析(PCA)方法对原输入空间进行重构,并根据各个主成分的贡献率来确定径向基网络结构,从而提高了网络的训练速度和评估能力。最后对某省实际运行的光纤保护通道进行风险评估来验证该方法的优越性。更多还原

【Abstract】 The risk evaluation of power system communication fiber-optic relay protection channels is investigated. The paper uses this approach of principal component analysis (PCA)-radial basic function neural network(RBFNN) to evaluate the power system fiber-optic relay protection channel. The proposed method avoids the complexity of influencing factor and difficulty when describing evaluation in models of mathematics. Concretely,the original input space is reconstructed by principal component analysis and the structure of the network is determined according to the contributions from the principal components respectively, so the ability of training speed and evaluation are improved.The effectiveness of the proposed algorithm is verified by the practical data for the fiber-optic relay protection channels.更多还原