Search

Network intrusion detection plays a critical role in cyberspace security. Most existing conventional detection methods mostly rely on manually-designed features to detect intrusion behaviours from large-scale flow data. Recent studies show that deep learning-based methods are effective for network intrusion detection due to the ability to learn discriminative features from data automatically. However, these models ignore the problem of the irregular time intervals between packets in a flow, causing the degradation of detection performance. To this end, we propose a Spatial and Temporal Aware Intrusion Detection model (STIDM). The proposed STIDM model first uses a one-dimensional Convolutional Neural Network (1D-CNN) to extract spatial features based on the nature of flow and packet. Then we design a Time and Length sensitive LSTM (TL-LSTM) method to learn richer temporal features from the irregular flows. The two parts are trained simultaneously to achieve global optimum. Through extensive experiments on the ISCX2012 dataset and the CICIDS2017 dataset, we demonstrate that STIDM outperforms state-of-the-art models.

Aiming at the measurement error of capacitance current for resonance grounding distribution network caused by damping resistance and the internal impedance of voltage transformer, a capacitance current measurement method for resonant grounding distribution network considering damping resistance was presented in this paper. The method is to inject a non-power-frequency signal into the distribution network through one of the two transformers of the voltage transformer inside the arc-suppression coil or zero-sequence voltage transformer of distribution network, and the returned signal is measured from another no-load voltage transformer, which eliminates the influence of the voltage transformer’s internal impedance. By establishing the zero-sequence equivalent circuit of distribution network, the capacitance current can be obtained accurately which eliminates the influence of damping resistance. Finally, a 10kV distribution network simulation model was established by PSCAD simulation software, the simulation analysis results showed the proposed method can accurately obtain the capacitance current of distribution network.

In order to accurately and effectively realize the arc-suppression coil tuning, it is urgent to improve the measuring accuracy of the capacitance to ground in resonant grounding system. Because the existing measuring method does not consider the damping resistance which is connected in the grounding loop of arc-suppression coil, a novel measuring method of capacitance to ground in resonant grounding system was proposed in this paper. First, a variable-frequency current signal is injected into the system by the zero-sequence voltage transformer of the distribution network. Then the returned voltage signal will be measured by the voltage transformer inside the arc-suppression coil. After establishing the variable-frequency equivalent circuit which considers the damping resistance, the capacitance to ground can be calculated accurately. Finally, the effectiveness of this method was proved by the PSCAD/EMTDC, which showed that this method has high precision, which eliminates the negative influence of damping resistance.

Aiming at suppressing three-phase unbalanced overvoltage in distribution network, a new suppression method was proposed in this paper. The PWM active inverter is used to inject a current with the amplitude and phase easily obtained into the neutral point of distribution network. The optimal injection current only need to record the amplitude and phase of the neutral point voltage at the initial moment of the three-phase unbalance, and takes it as a reference. After injecting the optimal injection current, the neutral point voltage can be suppressed to 0, i.e., the purpose of three-phase unbalanced overvoltage suppression can be achieved. Simulation analysis showed that it has the fast response, and it can effectively suppress the three-phase unbalanced overvoltage, which verified the feasibility and validity of the proposed method.

The experimental study of aerobic granular sludge in a sequencing batch reactor (SBR) was performed using synthetic domestic wastewater and ordinary flocculent activated sludge as seeding sludge. The conditions under which the successful cultivation of aerobic granular sludge was done are as follows: the temperature is room temperature, pH is about 7.0, aeration time is 150 ∼ 180min, sludge settling time is 9 ∼ 120 min, and aeration rate 0.3 ∼ 0.5 m3/h.. The aerobic granular sludge has yellow appearance, with a size of 1 ∼ 2 mm, SVI of 40 mL/g or so, and MLSS of 6515 mg/L. The average settling velocity is up to 18.42 m/h. Aerobic granular sludge contains large amounts of protozoa and metazoan which emerge in groups. The removal rates of COD, TN and TP is respectively up to 95%, 96% and 98%.

This paper is concerned with an optimal design of wastewater reuse for a University. The raw water and the capacity of wastewater reuse were determined according to the wastewater characteristic of a university; the treatment process was chosen by building a cost-benefit model; and the modeling test was performed about the operation effect of the proposed project and about the recommended water quality of reclaimed water. It is estimated that 3.4×105m3/a tap water and 6.1×105 Yuan/a will be saved after the project being put into operation, which shows that the proposed project will reduce the total quantity of pollutant efficiently and produce a great deal of social and economic benefits.

The purpose of this paper is to present a 2D depth-averaged model for simulating and examining flow patterns in channel bends. In particular, this paper proposes a 2D depth-averaged model that takes into account the influence of the secondary flow phenomenon through the calculation of the dispersion stresses arisen from the integration of the products of the discrepancy between the mean and the true velocity distributions. The k - ɛ turbulence model is used to compute turbulence stresses. As for the numerical solution procedure, the Finite Volume Method with the staggered grid and the SIMPLEC solution method are used to numerically solve the flow governing equations. Practical application of the model is illustrated by an example, and a fair agreement between the values measured and computed demonstrates the model's capabilities.

This paper is concerned with a mathematical model for simulating hydrodynamics of 2D river flow with groins based on the WENO scheme and the Finite Volume Method on unstructured grids. The time discretization uses the Runge-Kutta TVD scheme. By using the proposed model, we calculated the flow property with a groin, and obtained the flow velocity field distributions. The comparisons of the calculated results and the measured data show that the WENO scheme has higher accuracy and better stability, and has the ability to automatically capture shock waves, and may suppress the oscillations of numerical solution in the vicinity of discontinuities. This model can effectively simulate the hydrodynamics of 2D river flow with groins.

This article puts forward theoretical model on ISAR deception jamming by theoretical analysis, including space model and jamming signal model. Space model includes location relation and movement relation between radar, target and jamming platform, jamming signal model based on the relation between jamming signal and radar signal, distance migration of jamming signal, Doppler frequency of jamming signal, rotating feature of jamming signal and so on. Finally, correctness of model has been verified by simulation test.

This paper presents a formulation of slurry for Cu CMP, in which colloidal silica (SiO/sub 2/) was used as the abrasive particles. Since Cu is chemically active we tried to planarize Cu by "soft friction" using colloidal silica (size of 20nm), with the selected oxidizer in which metal ions aren't involved, and organic amine is used as the complex agent. Silicon sol (size of 20nm) is softer than Al/sub 2/O/sub 3/, has no contamination, and better dispersity and suspension with little scratching. Moreover, adding non-metal ions to the chelate can compress contamination of metal ions and adsorption of particles, so that the surface is smooth and bright.