In this paper a visual fusion analysis method for network security data based on time series is proposed. The correlation between network data is used for feature selection and consistency filtering. The original features are preserved while the redundant features are eliminated. The information entropy between network data features is proposed. Calculation

using the INTERACT method to achieve normalization processing; through the time series analysis method for visual fusion analysis of the processed network data

using the Chebyshev method to verify the network packets to determine whether there is intrusion behavior

to achieve network security data Visual fusion. The experimental results show that the detection rate can be improved by regulating the detection threshold. With the continuous expansion of data size

the detection rate of the proposed method is relatively stable

and the false alarm rate is lower than other methods.