Abstract:

Abstract: In spatial databases, spatio-temporal indexes based on R tree use minimum bounding box(MBB)to approximate spatio-temporal data to improve query efficiency, and the efficiency of indexing can be further improved by by clipping the bounding box.To address this problem, an improved spatio-temporal indexing method based on CBB was proposed, which firstly extended the method from two dimensions to the spatio-temporal dimension and obtained the possible clipping points by calculation, and recorded these points in the index for optimizing the clipping space of the bounding box of the index nodes, which could reduce the unnecessary child-node search in the query process.Then, the intersection of query box and the MBB of index node was analyzed, and the range query processing algorithm was further optimized, thus avoiding redundant comparison of clipping points in the query process.The experimental results showed that the indexing method could clip the space of the MBB of index nodes three times more than the original method, and could reduce the node computation by 40%, and could reduce the query time by 20%, which further could improve the query performance of spatial division tree-based spatio-temporal index.

Abstract: In the information fusion procedures of the traditional Beidou/INS integrated navigation systems, the correlation between the estimated sources was not considered, which could not guarantee the consistency of information fusion, a split covariance intersection filter(split CIF)-based information fusion algorithm was proposed.The algorithm separated the error covariance matrix into correlated part and independent part for processing, respectively.Consistent fusion estimation was obtained by measuring information and optimal filter gain updated in real time, which overcame the problem of the filter over-convergence caused by the fusion correlation data.in addition, in high precision integrated navigation, the lever arm error is an important error source.The accuracy of navigation positioning results could decrease severely for Beidou/INS integrated navigation systems.In this paper, the method of static lever arm measurement and error compensation for INS update information was used to reduce the influence of lever arm effect and further to improve the positioning accuracy.The simulation results showed that split CIF considering lever arm error compensation had good accuracy for lever arm estimation and high navigation accuracy.The maximum error accuracy of east velocity was 0.041 m/s, and the maximum accuracy of latitude error was 0.495 m.Compared with the compensated Kalman filter, it increased by 45.8% and 34.0%, and compared with EKF, the error accuracy increased by 55.1% and 10.0%, respectively.

Abstract: To locate the target indoor precisely, it was important to investigate on the Wi-Fi based indoor localization parameter, namely, time of arrival(TOA).The main weakness of primary TOA estimation methods were estimation bias, complex algorithm，and lack of universality.The new method had a clear physical meaning, and could pave a new way for the TOA estimation.According to the weakness of these traditional methods, a new TOA estimation algorithm was proposed in this paper.Based on sampling of waveform, the space-frequency-time searching algorithm, this algorithm calculated the phase difference between adjacent CSI calculation time span, and judged whether the CSI calculation time span could reach the edge of waveform, as to get the result of TOA estimation.Numerical results showed that the new method could achieve more than 58% submeter-level precision when the signal-to-noise ratio was above 25 dB.Moreover, the performance of the method upgraded as the bandwidth of the Wi-Fi signal expanded.The new method could overcome the weakness of traditional method, and pave a new way for the TOA estimation.

Abstract: Intelligent optimization algorithms could be divided into four categories: nature-like optimization algorithm, evolutionary algorithm, plant growth simulation algorithm，and swarm intelligence optimization algorithm.The swarm intelligence optimization algorithm was the most important type of algorithm.It played an important role in solving complex engineering problems, and together with image processing, fault detection, path planning, particle filtering, feature selection, production scheduling, intrusion detection, support vector machines, wireless sensors, neural network models, and got more extensive applications in other fields.In recent years, intelligent optimization algorithms such as bat algorithm, fruit fly optimization algorithm, whale optimization algorithm, salp swarm algorithm, and harris hawks optimization algorithm were widely used.Based on these five new swarm intelligence optimization algorithm, the model, characteristics, improvement strategies and application fields of the algorithm were reviewed.It analyzed the development opportunities and future trends it faced from theoretical investigations, improvement strategy and application studies, and provided a guidance on algorithm application.Findings showed that swarm intelligence optimization algorithm could perform well on many classic problems, but still should be expanded in the fields of multi-objective optimization, multi-constraint optimization, dynamic optimization, and mixed variable optimization.Effective parameter control of different groups of intelligent optimization algorithm in the face of various specific problems was still the focus of future studies.Co-evolution from populations, exploring more efficient hybrid methods and search strategies could be feasible solutions.

Abstract: Aiming at the problem of fewer tasks and low accuracy of recognition for motion imagination electroencephalogram(EEG)signals, in this paper a common space pattern(CSP)method based on wavelet packet decomposition(WPD)was proposed to extract the features of EEG signals.The long short term memory network was used to decode the EEG signals.Motor imagination signals were separated from blind sources by independent component analysis(ICA), and each channel EEG signal was divided into 8 groups by frequency using wavelet packet decomposition.The power value of each group of signals was calculated, 10 features were removed for classification by recurrence feature elimination(RFE).The selected signals were extracted by one-to-one common space pattern filters.The feature matrix were input into a long shortterm memory network for EEG decoding, and the classification results of 4 categories of motion imagination signals were obtained.The proposed method was used to verify the open data set of brain computer interface(BCI)competition(including four kinds of EEG signals: left hand imagination signal, right hand imagination signal, tongue imagination signal, and foot imagination signal).The recognition accuracy of three subjects was 90.28%, 94.25% and 96.55% respectively, and the average recognition accuracy could reach 93.69%.Compared with other feature extraction and classification methods, this method had a high classification accuracy.The decoded EEG signals were used to control the clockwise or counterclockwise movement of the space manipulator to achieve the purpose of grasping the target in the virtual space environment.

Abstract: Aiming at the problems of unbalanced data types and incomplete feature learning in deep learning intrusion detection, a neural network intrusion detection model based on the fusion of convolutional neural networks(CNN)and bidirectional gated recurrent unit(BiGRU)was proposed.The SMOTE-Tomek algorithm was used to balance the data set, the feature importance algorithm based on mean decrease impurity was used to realize feature selection; the CNN and BiGRU models used for feature fusion and attention mechanism was introduced for feature extraction, so as to improve the overall detection performance of the model.The intrusion detection data set CSE-CIC-IDS2018 was used for multi classification experiments, the model was compared with the classical single deep learning models.The experimental results showed that, firstly, in terms of data set balance, after being processed by SMOTE-Tomek algorithm, the recognition accuracy of DoS attacks-Slow HTTP Test class was improved from 0 to 34.66%, that of SQL Injection class was improved from 0 to 100%, and DDoS attack-LOIC-UDP, Brute Force-Web and Brute Force-XSS classes were improved by 5.22 percentage points, 6.55 percentage points and 35.71 percentage points respectively.It was proved that the balanced data set improved the recognition accuracy of a few classes significantly compared with the unprocessed data set.Secondly, in terms of the overall detection performance of the model, in the comparison of multi classification experiments, the overall classification accuracy, recall and F1 value of the model in this study were higher than those of several other single neural network models.The overall evaluation accuracy of each attack traffic category was about 2.10 percentage points higher than that of the highest LSTM model.The recall rate of the overall evaluation was about 1.50 percentage points higher than that of the highest LSTM model.Compared with the highest GRU model, the overall F1 value increased by about 1.97 percentage points.It was proved that the model had better detection effect.

Abstract: Due to the unbalanced proportion of abnormal traffic data in network abnormal traffic detection, the model could not fully learn rare attack traffic, which might affect the model training and detection accuracy.To solve this problem, a network intrusion detection model based on DBSCAN_GAN_XGBoost was proposed.When the model expanded rare attack samples, it focused on the noise samples that could more likely cause confusion in machine learning.Firstly, the DBSCAN algorithm was used to cluster the extracted rare attack data categories to generate one or more sub-clusters, and then the samples inside the cluster and the noise samples outside the cluster were extracted.Then, the generative adversarial network model was used to expand the extracted in-cluster samples and noise samples respectively, and to change the original sample proportion.Finally, the reconstructed data set was used to train the XGBoost algorithm based on decision tree classifier, and a complete the detection of abnormal network traffic data.UNSW-NB15 data set was used for comparative experiment, and the experimental results showed that the accuracy, and accuracy of DBSCAN_GAN_XGBoost model were 98.76% and 96.5% respectively, which were 15.63 percentage points and 19.60 percentage points higher than that before sample expansion, and effectively improved the detection accuracy of rare attack categories.

Abstract: In order to find out the regeneration effect of furfural extraction oil as asphalt regeneration agent on asphalt with different aging degrees, the rheological properties and temperature sensitivity of furfural extraction oil recycled asphalt were studied.By adding furfural extraction oil to asphalt with mild, moderate and severe aging degrees, the dynamic shear rheological tests and bending creep tests were used to study the influence of furfural extraction oil on the rheological properties and temperature sensitivity of aged asphalt, and the rheological properties and temperature sensitivity of recycled asphalt were evaluated with different indexes.The results showed that with the increase of furfural extraction oil content, the middle temperature fatigue resistance and low temperature crack resistance of aging asphalt were improved, and the high temperature deformation resistance was reduced.The performances of asphalt with different aging degree could be restored to the level of matrix asphalt under suitable furfural extraction oil content.The multiple stress creep test was more suitable for evaluating the high temperature deformation resistance of asphalt than the rutting factor.The evaluation results of the fatigue factor and linear scanning amplitude test on mesothermal fatigue resistance of asphalt were consistent.The temperature sensitivity of recycled asphalt had different performance in different temperature ranges.In the temperature range of 46-76 ℃, with the increase of furfural extraction oil content, the temperature sensitivity of three kinds of recycled asphalt was enhanced.In the temperature range of-18--6 ℃, the evaluation results of the creep rate index on the temperature sensitivity were better than those of the creep stiffness index.Temperature sensitivity of severe aging recycled asphalt increased with the increase of furfural extraction oil content, while the temperature sensitivity of mild and moderate aging recycled asphalt decreased.With the basic recovery of each index to the same as the virgin asphalt, the optimum content was determined as mild aging 4%, moderate aging 7%, severe aging 8%.

Abstract: Aiming to solve the problem of the stability of roadbed in the condition of rainfall infiltration in loess area, cement-soil mixing pile was adopted as the foundation reinforcement treatment method.Based on the foundation geotechnical test of loess and the unconfined compressive strength test results of cement soil modified with different amounts of fly ash and coal gangue, according to the M-C criterion and the load-settlement principle of roadbed in rainfall infiltration conditions, the roadbed model treated with cement soil mixing pile in rainfall infiltration was constructed by using the finite element analysis software ABAQUS.Combined with the numerical simulation results, the load settlement variation law of loess roadbed reinforced with cement-soil mixing piles with different rainfall duration, rainfall intensity and pile elastic modulus was analyzed; the bearing characteristics of loess roadbed were discussed; and the influence of rainfall infiltration on the stability of loess roadbed was explored.The research results showed that the cement-soil mixing pile had a significant strengthening effect on the vertical direction of the loess roadbed in the condition of rainfall infiltration, and the displacement of the roadbed slope foot could increase with the increase of rainfall intensity and rainfall duration.In the condition that the rainfall intensity was 2 mm/h and the rainfall duration was 18 h, the road pavement settlement decreased with the increase of pile elastic modulus.When the elastic modulus of the pile was 300 MPa, the cement-soil mixing pile had the best reinforcement effect on the loess roadbed.

Abstract: Using ABAQUS software, the finite element calculation model of square piers with tenon-grouting sleeve mixed connection was put forward by selecting material constitutive model and element type reasonably, dividing mesh and establishing boundary condition.The failure mode of concrete and the ultimate bearing capacity of concrete under axial load were compared with the test results, and the accuracy of finite element calculation was verified, the stress state of internal steel bar, steel pipe and grout sleeve was analyzed as the pressure increased.On this basis, taking the design parameters of CFST tenon(concrete-filled steel tube tenon)as the variables, the influence of the variables on the axial compression performance of the short pier of tenon-grouting sleeve hybrid connection was further studied, the results showed that the proposed finite element model could simulate the axial compression process of the square piers with tenon-grouting sleeve joint, the mean and variance of the ultimate bearing capacity calculated by finite element method and the test results were 1.036 and 0.023, respectively.Increasing the length and diameter of CFST tenon steel pipe could obviously increase the ultimate bearing capacity of the mortise-grouting sleeve connection bridge pier, the change of the thickness of CFST tenon steel tube had an increasing and decreasing trend for the ultimate bearing capacity of mortise-tenon-grout sleeve bridge pier, but it has little effect on the length and diameter of steel tube.In this study, a simplified formula for calculating the axial bearing capacity of square piers with mortise-tenon-grouted sleeve connection was presented.The calculated results were slightly lower than the results of tests and finite element analysis, and the average error between the calculated results and the experimental results was 9.01%, the error was 4.05%, and the error was small.It could be used to calculate the axial bearing capacity of this type of pier.

Abstract: This study aimed to solve the problem of deep purification of high concentration phosphorus wastewater and to explore the phosphorus removal potential of portland blast furnace slag cement.The adsorption of the wastewater containing 150 mg/L TP on portland blast furnace slag cement was discussed.Combined with scanning electron microscopy(SEM)and X-ray diffraction(XRD), the surface morphology and crystal structure of portland blast furnace slag cement before and after phosphorus adsorption were characterized and analyzed.The results of adsorption kinetics experiments showed that portland blast furnace slag cement as a adsorbent could treat high phosphorus-containing wastewater, removal rate up to 90.3%, adsorption capacity up to 27 mg/g, the adsorption kinetic process could be well explained by using pseudo-second-order kinetic model and intra-particle diffusion model, the dephosphorization rate is mainly limited by the chemical reaction rate, compared with surface adsorption, intra-particle diffusion is the decisive step in the adsorption process.In addition, isotherm studies showed that phosphorus adsorption on portland blast furnace slag cement fitted the Freundlich and Langmuir model, with the correlation coefficients over 0.97.The results indicated that both single-layer adsorption and multi-layer adsorption occurred in portland blast furnace slag cement, and the non-uniform active sites on the surface were slightly stronger than the uniform active sites, the maximum phosphorus adsorption capacity determined by Langmuir isotherm plot was 47 mg/g.Through orthogonal experiments, the optimal reaction conditions were as follows: pH is 7, initial phosphorus concentration is 150 mg/L, dosage is 5 g/L, contact time is 18 h.The portland blast furnace slag cement can effectively neutralize the excess acid and alkali, and has good buffering characteristics, which may be related to the protonation and deprotonation of the metal ions, the ligand exchange in chemisorption was the main mechanism of phosphorus removal.Portland blast furnace slag cement is suitable for high concentration of phosphorus containing wastewater, also suitable for pH unstable phosphorus containing wastewater.

Abstract: This study aimed to explore the fire extinguishing effect of different particle size NaHCO3 powder, to compare the synergistic fire extinguishing effect between it and different typical gas fire extinguishing agents, to obtain the best gas-solid ratio of synergistic effect, and to provide data support for the research of gas-solid synergistic fire extinguishing.Taking 4 μm, 10 μm, and 48 μm NaHCO3 powders with three different particle sizes as the research object, according to the scanning electron microscope observation of different particle sizes of NaHCO3, the uniform distribution of the powder particles was determined, and the extinguishing was carried out based on the cup burner device.In the methane flame experiment, the influence of the particle size on the powder fire extinguishing effect was analyzed according to the critical fire extinguishing concentration relationship obtained from the experiment, and the 4 μm NaHCO3 powder with the best fire extinguishing effect was selected and the typical gas fire extinguishing agent CO2, N2, He, Ar was used for gas solid-solid synergy experiment to obtain the critical fire-extinguishing concentration of NaHCO3 powder at different typical gas fire extinguishing agent concentrations, so as to calculate the synergy factor to quantitatively evaluate the gas-solid synergistic fire extinguishing effect, and select the gas-solid ratio and typical gas fire extinguishing agent with the best synergistic effect.Experimental results showed that the smaller the average particle size of the powder, the better the fire extinguishing effect.The order of fire extinguishing efficiency was 4 μm NaHCO3 >10 μm NaHCO3 >48 μm NaHCO3.The fire extinguishing efficiency of 4 μm NaHCO3 powder was about 2.2 times of 10 μm NaHCO3; 4 μm NaHCO3 powder and gas fire extinguishing agent had a certain positive synergistic effect.With the increase of the volume fraction of gas fire extinguishing agent, the critical fire extinguishing concentration of NaHCO3 powder decreased.In the atmosphere, there was a certain similarity in the downward trend of the critical fire-extinguishing concentration.When the CO2, N2, He, and Ar gas concentrations were 7.1%, 6.4%, 8.5%, 10.1%, the synergistic effect with 4 μm NaHCO3 powder reached its peak, and the corresponding synergy factors were 0.58, 0.62, 0.69, 0.76; Among the typical gases, CO2 had the best synergistic effect.

Abstract: Polyethylenimine/Salix(PEI-SP)was prepared by epichlorohydrin crosslinking for application in the adsorption of acid chrome blue K(ACBK)from aqueous solution.The adsorption kinetic models and the adsorption isotherm models were used to fit the obtained experimental data, respectively.The results showed that the kinetic process was best predicted by the Elovich model, which indicated the adsorption process was an inhomogeneous chemisorption process and the adsorption rate was controlled by film diffusion.The equilibrium isotherm was best fitted with the Koble-Corrigan model, which indicated that the adsorption of ACBK by PEI-SP was mainly monolayer adsorption, while accompanied by multi-molecular layer adsorption.The maximum adsorption capacity of PEI-SP calculated by Langmuir model for ACBK was 0.428 mmol/g at 298 K.As could be seen from the influence of pH value on the adsorption capacity, the adsorption capacity of PEI-SP on ACBK gradually decreased with the increase of pH value.When pH<6.78, ACBK was mainly adsorbed on the surface of PEI-SP by electrostatic attraction.Combining the results of the infrared analysis and the effects of pH values on the adsorption capacity, the adsorption mechanism of ACBK on PEI-SP were electrostatic interaction, hydrogen bonding, and π—π stacking interactions.The results of desorption and regeneration experiments showed that the adsorbent could retain its adsorption capacity after three cycles operations, showing excellent regenerability.It was implied that PEI-SP has a promising application in the removal of anionic dyes from wastewater.

Abstract: In this study, the discharge efficiency and mass transfer effect were improved with the Ar/O2 mixed gas as discharge gas, a pulse power supply as driving power, and a specially designed microporous ceramic electrode.Discharge characteristics underwater were systematically studied by the analysis of the lowest excitation voltage, emission spectrum, active substance generation(such as·OH、H2O2、O3), and Indigo Carmine degradation characteristics.Experimental results showed that this system could achieve a large-area uniform discharge underwater with driving voltage as low as 2.1 kV, the O2 proportion had the greatest effect on the excitation voltage of the discharge.Although the attempt to find the characteristic spectrum of·OH failed, active substances such as H and O in excited state were detected in the spectral analysis results.The increase of frequency and voltage was conducive to the generation of active substances, such as OH, H2O2, O3, and the dye decolorization efficiency.The increase of gas flow was in favor of the H2O2 and O3 generation, but adverse to the·OH generation, and had little effect on the dye decolorization efficiency.As the O2 proportion increased, the generation amount of O3 increased gradually, but the·OH, H2O2 generation, as well as the decolorization efficiency increased firstly and then decreased.Discharge conditions and cost problems were comprehensively considered, the optimum condition of this discharge system was as follow: gas flow 200 sccm(Ar 80%，O2 20%), frequency 2 kHz, voltage 5 kV, pulse width 1 μs.The average decolorization rate of Indigo Carmine reached 97.8% in 10 minutes in this ideal discharge condition.

Abstract: According to the geometry and kinematics theory of the cycloid pin wheel of the RV reducer, the equivalent radius of curvature, entrainment speed and load per unit length of the cycloid pin wheel transmission at different meshing positions were analyzed.Based on the Ostwald model of grease, a linear contact grease lubrication elastohydrodynamic analysis model was established, and the influence of load, entrainment speed and rheological index on the grease film thickness and pressure distribution was analyzed.Numerical analysis of grease lubrication at different meshing positions of the cycloidal pinwheel was carried out, and the shape and pressure distribution of the grease film at the discrete points of meshing are obtained, and then the instantaneous lubrication state of the needle teeth during the meshing process from the root of the cycloidal gear to the tooth tip was obtained by calculating the film thickness ratio, and the poor lubrication area of the cycloidal gear was determined.The results showed that the film thickness and pressure distribution characteristics of grease lubrication and their changes with load and entrainment speed were similar to those of oil lubrication, but from a quantitative point of view, the film thickness of oil lubrication was higher than that of oil lubrication in the same conditions.In the RV reducer cycloid pin wheel drive, as the pin teeth mesh from the tooth root to the tooth tip, the equivalent radius of curvature decreased rapidly and then changed slowly.The entrainment speed first decreases and then slowly increased, and the load per unit length increased rapidly and then slowly decreased.The change in equivalent radius of curvature, entrainment speed and load per unit length also changed the lubrication performance of different meshing points.When the needle teeth mesh at the root of the cycloidal gear, the thickness of the grease film was larger.The lubrication performance from the root of the cycloidal gear to the top of the tooth gradually changed from good to worse, and the top of the cycloidal gear was a poorly lubricated area.The thickness of the grease film increased with the increase of the rheological index.By increasing the rheological index of the grease, the lubrication state of the cycloidal pin wheel transmission could be effectively improved, and the life of the cycloidal wheel could be improved.

Abstract: In order to effectively solve the vehicle-to-vehicle collision problem in the lane change of adjacent vehicles, which could improve the active safety of intelligent vehicles, a lane change trajectory prediction model satisfying both short-term prediction and long-term prediction was constructed by using the weighted fusion of CTRA motion model and quintic polynomial model.Based on this, the lane change safety distance model was established, and an appropriate active collision avoidance algorithm was designed.The results of PreScan/Simulink co-simulation showed that the prediction accuracy of the trajectory weighted fusion model was significantly higher than that of the two basic models, and the absolute deviation of the short-time and long-time model was 0.09 m and 0.18 m in the established lane change simulation condition.The design of safety distance showed that when D≤Dwmin or D≥Dwmax, the vehicle had no collision risk, and when Dwmin