CAJ | 학술논문

针对电力行业内传统网络安全预测无法全面反映系统整体状况，预测精度不高的缺点，提出一种网络安全预测方法。首先对网络安全事件进行分析，采用层次分析法构建网络安全指标体系，并对样本数据进行处理，构造异常值的分布序列，而后采用灰色方法进行预测模型建模，对预测结果运用神经网络方法进行残差修正，从而实现提高预测精度的目的。通过仿真实验，表明基于残差改进的灰色模型的网络安全预测方法的可行性和有效性。
침대전역행업내전통망락안전예측무법전면반영계통정체상황，예측정도불고적결점，제출일충망락안전예측방법。수선대망락안전사건진행분석，채용층차분석법구건망락안전지표체계，병대양본수거진행처리，구조이상치적분포서렬，이후채용회색방법진행예측모형건모，대예측결과운용신경망락방법진행잔차수정，종이실현제고예측정도적목적。통과방진실험，표명기우잔차개진적회색모형적망락안전예측방법적가행성화유효성。
The paper suggests a new forecasting model for the network security-related problems in the power industry to remedy the shortcomings of the traditional ones which fail to reflect the industry’s overall conditions and cannot accurately predict. The sample data is collected by analyzing the events concerning the network security. Then AHP (analytic hierarchy process) is applied to set up an indicator system to evaluate those data and form a sequential distribution of exceptional values. Based upon that, GM (Grey Model) is introduced to comprehensively predict the conditions of the industry’s information security, and then the prediction results are modified by using artificial neural network method. The simulating tests have also been carried out to prove that the proposed model with improved GM as the basis is viable and valid.