Search results

The purpose of this paper is to solve the shortage of the existing methods for the prediction of network security situations (NSS). Because the conventional methods for the prediction of NSS, such as support vector machine, particle swarm optimization, etc., lack accuracy, robustness and efficiency, in this study, the authors propose a new method for the prediction of NSS based on recurrent neural network (RNN) with gated recurrent unit.

This method extracts internal and external information features from the original time-series network data for the first time. Then, the extracted features are applied to the deep RNN model for training and validation. After iteration and optimization, the accuracy of predictions of NSS will be obtained by the well-trained model, and the model is robust for the unstable network data.

Experiments on bench marked data set show that the proposed method obtains more accurate and robust prediction results than conventional models. Although the deep RNN models need more time consumption for training, they guarantee the accuracy and robustness of prediction in return for validation.

In the prediction of NSS time-series data, the proposed internal and external information features are well described the original data, and the employment of deep RNN model will outperform the state-of-the-arts models.

In the measurement of liquid density and viscosity, the change of resonance parameters due to the parasitic parallel capacitance of resonator affects the measurement accuracy. To improve the accuracy, a method was proposed to compensate the parasitic parallel capacitance of resonator by adding an electrode.

The new electrode (compensation electrode) was added into resonant sensor to make compensation capacitance. The closer the compensation capacitance was to the parasitic parallel capacitance, the better compensation was. The structural parameters of resonant sensor with the compensation electrode were determined by the simulation and experiment.

The effect of this method was examined by the experiment. The relative errors of density and viscosity were less than 0.15, 0.5 % and standard deviations were less than 0.0004 g/cm³ and 0.005 mPas, respectively.

The experimental results show that this method is valuable for the parasitic parallel capacitance compensation of immersed resonant sensor.

This paper has not been published in other journals.