Abstract:The content and distribution of nitrate and ammonium in the unsaturated soil were analyzed in this study, considering four kinds of typical planting patterns on both sides of the agricultural irrigation area of Kuihe River in Tongshan District of Xuzhou City. The analysis results of the soil leaching liquor, which was obtained from the situ soil sample, show that the content of nitrate vary from 3. 8 to 128. 1 mg/kg with an average of 33. 0 mg/kg and the content of ammonium vary from 2. 3 to 5. 7 mg/kg with an average of 3. 8 mg/kg. The annual fertilization and sewage irrigation is the main source of nitrogen in the unsaturated soil. Meanwhile, the distribution of nitrogen in the soil is controlled by the condition of fertilization and irrigation for different planting patterns. Furthermore, the nitrogen is recharged by the surface irrigation and fertilizer, and mainly migrates in the direction of down-gradient of soil column. The ammonium is absorbed in the process of infiltration and as a general rule, the content of ammonium is related to the number of clay particles. However, the content of ammonium is still at a high level in the planting area of rice and cabbage with annual irrigation, even though the number of clay particles is small near the soil surface.

Abstract:In order to quantitatively characterize the non-uniformity of precipitation in the Huaihe River Basin, the centroid time of annual precipitation was established to generalize the time distribution of precipitation process. The overall pattern and evolution characteristics of precipitation concentration period in the past 55 years was analyzed from three aspects: time domain, space domain and frequency domain. The results show that: (a)the first main period of the centroid time of annual precipitation is 28 years in all selected stations, the mutation mostly happened in the southern area of the upper reaches of Huaihe River around from the late 1970s to the mid or late 1980s, and the time series of centroid time of annual precipitation has an increasing trend but not significant, which indicates that there was a slight backward in most areas of the watershed. (b)The centroid time of annual precipitation has a significant backward trend with the increase of the latitude, and the difference is about 20 days from the south to the north. The change of centroid time of annual precipitation has no correlation with the longitude, which explains that no obvious difference exists between the east and the west. (c)The centroid time of storm and heavy rain is generally later than the other grade of annual precipitation, to which the centroid time of annual precipitation is sensible.

Abstract:In order to deal the problem that many scholars lack the concept of catchment interval when modeling the waterlogging point, the method that can extract catchment characteristics of waterlogged points in the gentle rolling area of city has been established based on DEM and traditional D8 method to determine the flow direction and the filling of depressions. The depression, where the waterlogged points are located, was regarded as a “true depression”, which was not involved in the calculation of cycle filling. The DEM data of 5 m resolution for the Jiaonan District of Qingdao City was tested by the above method and 15 of the 18 waterlogging points were accurately identified, where the hit rate was 83%. In addition, the corresponding catchment characteristics were also extracted, and the water accumulation degree as well as the risk grade of each waterlogged point was analyzed by using the method. The No. 11 waterlogged point at Xiqueshan road is the most dangerous and potentially harmful. This method can better extract the catchment characteristics of waterlogging points in the urban area, exhibiting a reference value for the city’s flood control and disaster mitigation.

Abstract:The Fangbian Reservoir in the Lishui district of Nanjing City was selected as the study object. The ArcGIS technology was used in the watershed subdivision of catchment area in Fangbian Reservoir, and then combined with the remote sensing technology, this study simulated the pollution intensity and spatial distribution to determine the main pollution source of inflow stream and the intensity distribution of pollution source. After that, the simulating results of ArcGIS has been validated by the actually monitored water quality. The results showed that the catchment area of inflow stream and spatial distribution of pollution source intensity can be directly reflected by the ArcGIS technology. Besides, the difference of pollution emission amount calculated between the ArcGIS simulation and the surveyed data was less than 10%.

Abstract:In order to solve the tracing problem of pollutant source in the subject of water environment, a backtracking model was built in this study to discern the source location and the emission time, based on the convection-diffusion equation and the genetic algorithm. By further introducing the decoupling method, the mass quality was also obtained. A concept of ‘weigh coefficient’ was introduced to improve the traditional genetic algorithm for high computational efficiency by controlling the evolutionary trend of population and optimizing the quality of population. Compared with the traditional genetic algorithm, the improved one makes the backtracking model more efficient and stable.

Abstract:According to the numerical model of moisture-heat-stress coupled field, the finite element algorithm for soilbags was built, and then one canal engineering project in the cold area has been analyzed to obtain the distributions of temperature field, moisture field and displacement within two years after the completion of project. The results show that the canal slope encounters significant frost heaving and melting deformation during the freezing and thawing process, and both the frost heaving and the melting deformation are irreversible. The geotextile treated channel has a good performance in preventing frost heaving of frozen soil slope. It also has the advantage of small thermal conductivity that can effectively reduce the influence of atmosphere temperature on the soil temperature inside canal slope, so as to reduce the possibility of frost heaving and melting settlement. At the same time, soilbags can suppress the rise of capillary water and film water and reduce the water immigration of the geotextile bag layer and the lower part, so as to maintain a more stable water content within the channel to reduce the frost heaving of canal surface. In addition, the geotextile bag has a certain strength, and under the action of its tension stress, the frost heaving can be suppressed, thereby reducing the damage of frost heaving on the channel lining. At last, with the treated surface of channel using the geotextile bag, the internal temperature of slope can quickly reach a steady state, and after 2 years, a more stable and perennial frozen layer is formed below 2 m of the ground.

Abstract:In order to simulate the anomalous vertical profile of suspended sediment in the Jingjiang Reach, three equations, including the equilibrium equation by Rouse, the non-equilibrium equation by Han Qiwei and the fractional advection-diffusion equation, were adopted to describe the vertical distribution of unsaturated suspended sediment. The results showed that the equilibrium equation by Rouse can not describe the vertical distribution of sand content in the suspended sediment. Comparing the calculated data with the measured data, Han’s Equation performed well in the upper part, but underestimated the value near the riverbed. The fractional advection-diffusion equation can give a good description of the vertical distribution of concentration in channel with unsaturated sediment regime, when the new fractional parameters have reasonable values.

Abstract:To systemically study the effect of varying slope coefficients in the upstream side on the discharge coefficient at free overflow over a broken-line practical weir with vertical downstream face and without lateral contraction, different slope coefficients of upstream side, including 0, 0. 3, 0. 5, 0. 8, 1. 0, 1. 5 and 2. 0, are simulated by the volume of fraction(VOF)method and the RNG k-ε turbulence model. Numerical results show that the rising of total head H₀ reduces the curvature of free surface profile and increases the discharge coefficient. Meanwhile, the influence of slopes in upstream side on the discharge coefficient decreases with H₀. Furthermore, a discharge coefficient formula including the variable slope coefficient of upstream side is derived based on the method of non-linear regression, in which as the slope coefficient of upstream side increases, the discharge coefficient increases firstly and then decreases for a given H₀.

Abstract:The influence of floating structure on the downstream flowing structure in the flowing water was studied by using a physical model test. By comparing the flowing structures under different aspect ratios and water levels, the flowing velocity distribution, non-uniform coefficient and the length of recirculation zone of the characteristic cross section were analyzed. The results show that the influence of aspect ratios on the velocity distribution and the non-uniform coefficient is not obvious, while the water-level difference has a great influence and it shows an increasing trend with the increase of water level. At the same time, the length of recirculation zone is more sensitive to the aspect ratio and water level. In the practical engineering, attentions should be paid to the engineering problems caused by the change of water level. This study can provide references for the design of floating structure as well as the change rule of downstream flowing structure in current during the process of running.

Abstract:In order to avoid the influence of human behaviors in the data records of construction temperature and lift thickness, a portable detecting device has been developed with the functions of construction temperature measuring, lift thickness measuring, and simultaneous positioning. In the meantime, for the implementation of real-time visual management of the asphalt pavement quality, a three-dimensional highway construction model and a visual integration model of monitoring information were built, and then under the network environment, a real-time visualization system for the quality management of asphalt pavement was developed based on the portable detecting device. This system was applied in a highway project with asphalts pavement in Tianjin, and the qualified rates of the most measurements, except lift thickness and initial compacting temperature in several points, were up to 100%. The results indicate that the asphalt pavement temperatures and the lift-thickness of project were under good control. This system can improve the management efficiency of asphalts pavement quality, and provide a new technical approach for the quality control of road construction.

Abstract:Cohesive sediment is an important component of the aquatic environments, which must be monitored and managed for environment, engineering, and human health reasons. The stability of cohesive sediments in an aquatic environment dependent on the balance between hydrodynamic forces that cause erosion and the forces within the sediment that resist it. And the stability of cohesive sediments is described by the critical shear stress. At present, existing models for the critical shear stress of the surface erosion of cohesive sediments is lacking a consideration of the aggregate size distribution in cohesive sediments. This study constructed the fractal model of the cohesive aggregate structure in cohesive sediments. Meanwhile, the relationship is built for the critical shear stress as functions of the median diameter and excess density from the fractal model for cohesive aggregates. It can be found that the critical shear stress of cohesive sediments is a power function of the median diameter and excess density. The proposed relations from the fractal model for aggregates are validated by a bulk number of experimental data and a new method is proposed to determine the critical shear stress.

Abstract:The 3D laser scanning technology was applied to the deformation monitoring of subway tunnel, and for the pre-processing problem of point cloud data, a method of the point clouds filtering with grid handling was studied based on the wavelet analysis. Combining with the scanned point cloud data and its characteristics of Nanjing subway tunnel, this paper analyzed and selected a rational interpolation method to implement the grid processing of the cloud point data. The wavelet transform was used to detect the abnormal values of grid data, and the whole grid data was processed with the threshold de-noising method of wavelet. At last, this study packaged and compared the filtered point cloud data, indicating a very good filtering effect.

Abstract:One application challenge of the neural networks is the training, the essence of which is an iterative optimization process involving numerous data. The training process requires significant computing power and efficient searching methods for optimal solutions. To meet the requirement, a parallel BFGS quasi-Newton training algorithm based on GPU is proposed in this paper. To maximize the parallelism, the BFGS quasi-Newton algorithm is divided into different function modules, and each module is designed with a specific parallelism regarding its characteristics. In addition, the processing and memory resources of GPU are fully utilized to achieve a better parallelization. Experimental results show that the GPU implementation accelerates the neural network training by up to 80 times compared to the CPU implementation for the complicated neural network structure, while the speed up ratio is up to 430 for the modelling test of microwave device, where the training error is about 1%.

Abstract:The conventional algorithm can not deal with the deviant attitude of different experts in the evaluation of information security. At the same time, it adopts shallow machine learning models and mishandled the deviation accumulating problem. In the study, we proposed a novel evaluation algorithm of information security based on the fuzzy adjustment feature matrices and the deep time sequence model. In the proposed algorithm, the triangle fuzzy function was applied to build the expert’s evaluation indexes. Then, the improved weight DS theory of evidence was used to adjust the indexes. After that, the loss matrix and possibility matrix are constructed. The information security was evaluated by a deep neural model in proposed model. In the simulations of MIT dataset, we explored whether the feature matrices will cope with the conflicts of experts or not, and the accuracy, performance and robustness of estimation were also checked. It can be found that the proposed algorithm has a better ability to evaluate the fuzzy system, and the ability for coping with the conflicts of experts is also better, achieving a better accuracy and robustness with the reservation of performance.