论文部分内容阅读
提出一种基于动态灰色马尔科夫模型(dynamic gray Markov model,DGMM)的智能变电站网络流量预测方法.根据灰色预测理论对网络流量进行初步预测;针对初步预测误差,利用K-means聚类算法实现状态的划分,建立马尔科夫模型对预测误差进行动态校正;对K-means中的K值进行了讨论.利用真实环境下采集到的流量数据对所提方法进行测试.结果表明,该模型的平均预测误差仅有4%,能够满足预测要求,为智能变电站网络性能分析预测、网络故障和病毒入侵预警提供决策依据.
This paper presents a dynamic gray Markov model (DGMM) based network traffic forecasting method for smart substations. According to the gray forecasting theory, the network traffic is forecasted. For the initial forecasting error, K-means clustering algorithm The state is divided and the Markov model is established to dynamically correct the prediction error. The K value in K-means is discussed, and the real-time traffic data is used to test the proposed method. The results show that the model’s The average prediction error is only 4%, which can meet the prediction requirements and provide decision-making basis for intelligent substation network performance analysis and prediction, network failure and virus intrusion warning.