基于IABC和聚类优化RBF神经网络的电力信息网络安全态势评估

为提高电力信息网络安全态势评估的精度，提出一种基于改进人工蜂群（IABC）算法和密度峰值聚类（DPC）算法优化径向基函数（RBF）神经网络的电力信息网络安全态势评估方法。首先，引入改进密度峰值聚类（IDPC）算法对人工蜂群（ABC）算法的种群空间多样性进行聚类分析，重新定义个体更新机制以提高算法的全局搜索能力。然后，构建分类RBF神经网络安全态势评估模型，利用IDPC算法对输入指标数据进行聚类分析，采用IABC算法对分类拓扑结构和参数学习过程进行优化，得到输入评估指标与输出安全态势值的最佳映射关系。最后，通过实例仿真证明所提方法的有效性。

Security Situation Assessment of Power Information Network Based on IABC & Clustering Optimized RBF Neural Network

In order to improve the security assessment accuracy of power information network security situation，a power information network security situation evaluation method based on improved artificial bee colony algorithm （IABC） and density peak clustering （DPC） algorithm is proposed. Firstly，the improved density peak clustering （IDPC） is introduced to cluster the spatial diversity of artificial bee colony algorithm （ABC）， and the individual update mechanism is redefined to improve the global search ability of the algorithm. Then， the classified RBF neural network security situation assessment model is constructed， the input index data is clustered and analyzed by IDPC algorithm，and the classification topology and parameter learning process are optimized by IABC， so as to obtain the best mapping relationship between the input assessment index and the output security situation value. Finally，example simulation is given to demonstrate the effectiveness of the proposed method.