Comparative study on the performance of ConvLSTM and ConvGRU in classification problems-taking early warning of short-duration heavy rainfall as an example

卷积长短期记忆单元ConvLSTM和卷积门控循环单元ConvGRU是两种广泛应用的深度学习单元,通过将循环机制与卷积运算相结合,常常用于时空序列的预测.为了明确上述两种模型的收敛速度和分类能力,需要使用相同的模型架构对相同的分类问题进行预测.本研究将北京短时强降水区级预警问题看作深度学习中的二分类问题,使用京津冀雷达网的组合反射率数据和北京区域内的自动气象站降雨数据进行深度学习模型的训练和评估.结果表明,ConvGRU的收敛速度比ConvLSTM快约25%.ConvLSTM和ConvGRU的预警性能随地区,时间,降雨强度的变化趋势相似,但大部分ConvLSTM的得分较高,少数情况下ConvGRU的得分较高.

Spatiotemporal features and vertical structures of four types of mesoscale eddies in the Kuroshio Extension region

Except for conventional mesoscale eddies,there are also abundant warm cyclonic eddies(WCEs)and cold anticyclonic eddies(CAEs)in the global ocean.Based on the global mesoscale eddy trajectory atlas product,satellite altimetric and remote sensing datasets,and three-dimensional temperature/salinity dataset,spatiotemporal features of WCEs and CAEs are compared with traditional cold cyclonic eddies and warm anticyclonic eddies in the Kuroshio Extension(KE;28°−43°N,140°−170°E)region.Characteristics of abnormal eddies like radius,amplitude,eddy kinetic energy,and proportion in all eddies behave in significant asymmetry on the north and south sides of the KE jet.Unlike eddies in the general sense,temporal feature analysis reveals that it is more favorable to the formation and maintenance of WCEs and CAEs in summer and autumn,while winter is the opposite.The spatiotemporal variation of abnormal eddies is likely because the marine environment varying with time and space.Statistically,proportion of abnormal eddies increases rapidly in decaying stage during the whole eddy lifespan,resulting in smaller average radius,amplitude,sea surface temperature anomaly and sea surface height anomaly compared to normal ones.The three-dimensional composite structures for four types of eddies expose that the difference between abnormal and conventional eddies is not just limited to the sea surface,but also exists within the water below the sea surface.Vertical structures also indicate that the anomalous temperature signal is confined in the water from the sea surface to layers at about 30 m in the KE region.

Comparison and Verification of Coherent Doppler Wind Lidar and Radiosonde Data in the Beijing Urban Area

As a new type of wind field detection equipment, coherent Doppler wind lidar(CDWL) still needs more relevant observation experiments to compare and verify whether it can achieve the accuracy and precision of traditional observation equipment in urban areas. In this experiment, a self-developed CDWL provided four months of observations in the southern Beijing area. After the data acquisition time and height match, the wind profile data obtained based on a Doppler beam swinging(DBS) five-beam inversion algorithm were compared with radiosonde data released from the same location. The standard deviation(SD) of wind speed is 0.8 m s^(–1), and the coefficient of determination R~2 is 0.95. The SD of the wind direction is 17.7° with an R~2 of 0.96. Below the height of the roughness sublayer(about 400 m), the error in wind speed and wind direction is significantly greater than the error above the height of the boundary layer(about 1500 m). For the case of wind speeds less than 4 m s^(–1), the error of wind direction is more significant and is affected by the distribution of surrounding buildings. Averaging at different height levels using suitable time windows can effectively reduce the effects of turbulence and thus reduce the error caused by the different measurement methods of the two devices.

A Framework for Driver DrowsinessMonitoring Using a Convolutional Neural Network and the Internet of Things

One of the major causes of road accidents is sleepy drivers.Such accidents typically result in fatalities and financial losses and disadvantage other road users.Numerous studies have been conducted to identify the driver’s sleepiness and integrate it into a warning system.Most studies have examined how the mouth and eyelids move.However,this limits the system’s ability to identify drowsiness traits.Therefore,this study designed an Accident Detection Framework(RPK)that could be used to reduce road accidents due to sleepiness and detect the location of accidents.The drowsiness detectionmodel used three facial parameters:Yawning,closed eyes(blinking),and an upright head position.This model used a Convolutional Neural Network(CNN)consisting of two phases.The initial phase involves video processing and facial landmark coordinate detection.The second phase involves developing the extraction of frame-based features using normalization methods.All these phases used OpenCV and TensorFlow.The dataset contained 5017 images with 874 open eyes images,850 closed eyes images,723 open-mouth images,725 closed-mouth images,761 sleepy-head images,and 1084 non-sleepy head images.The dataset of 5017 images was divided into the training set with 4505 images and the testing set with 512 images,with a ratio of 90:10.The results showed that the RPK design could detect sleepiness by using deep learning techniques with high accuracy on all three parameters;namely 98%for eye blinking,96%for mouth yawning,and 97%for head movement.Overall,the test results have provided an overview of how the developed RPK prototype can accurately identify drowsy drivers.These findings will have a significant impact on the improvement of road users’safety and mobility.

Cost Optimization of Steel Beam-to-Column Connections using AVOA

The joint-bolt-African Vulture optimization algorithm(AVOA)model is proposed for the design of building connections to improve the stability of steel beam-to-column connections.For this algorithm,the type of steel is first determined,and the number of bolts needed by the corresponding steel type is referenced in Eurocode 3.Then,the bearing capacity of the joint can be calculated.The joint-bolt-AVOA model is established by substituting the bolt number required by the steel into the algorithm to obtain the optimal bolt number required while ensuring joint stability.The results show that the number of bolts required by the joint-bolt-AVOA model based on the stability of steel is lower than that calculated by Eurocode 3.Therefore,AVOA can effectively optimize the number of bolts needed in building connections and save resources.

Parameter optimization of microwave sintering porous Ti-23%Nb shape memory alloys for biomedical applications

Porous Ti-23%Nb(mole fraction)shape memory alloys(SMAs)were prepared successfully by microwave sintering with excellent outer finishing(without space holder).The effects of microwave-sintering on the microstructure,phase composition,phase-transformation temperature,mechanical properties and shape-memory effect were investigated.The results show that the density and size of porosity vary based on the sintering time and temperature,in which the smallest size and the most uniform pore shape are exhibited with Ti-23%Nb SMA after being sintered at 900°C for 30 min.The microstructure of porous Ti-Nb SMA consists of predominantα',α,andβphases in needle-like and plate-like morphologies,and their volume fractions vary based on the sintering time and temperature.Theβphase represents the largest phase due to the higher content ofβstabilizer element with little intensities ofαandα'phases.The highest ultimate strength and its strain are indicated for the sample sintered at 900°C for 30 min,while the best superelasticity is for the sample sintered at 1200°C for 30 min.The low-elastic modulus enables these alloys to avoid the problem of“stress shielding”.Therefore,microwave heating can be employed to sinter Ti-alloys for biomedical applications and improve the mechanical properties of these alloys.

Behavioral response of tilapia (Oreochromis niloticus) to acute ammonia stress monitored by computer vision

The behavioral responses of a tilapia (Oreochromis niloticus) school to low (0.13 mg/L), moderate (0.79 mg/L) and high (2.65 mg/L) levels of unionized ammonia (UIA) concentration were monitored using a computer vision system. The swimming activity and geometrical parameters such as location of the gravity center and distribution of the fish school were calculated continuously. These behavioral parameters of tilapia school responded sensitively to moderate and high UIA concen-tration. Under high UIA concentration the fish activity showed a significant increase (P<0.05), exhibiting an avoidance reaction to high ammonia condition, and then decreased gradually. Under moderate and high UIA concentration the school’s vertical location had significantly large fluctuation (P<0.05) with the school moving up to the water surface then down to the bottom of the aquarium alternately and tending to crowd together. After several hours’ exposure to high UIA level, the school finally stayed at the aquarium bottom. These observations indicate that alterations in fish behavior under acute stress can provide important in-formation useful in predicting the stress.

Microstructural evolution during semisolid processing of Al-Si-Cu alloy with different Mg contents

A series of Al-6Si-3Cu-(0.3-2)Mg alloys were produced by a conventional casting process.Cooling slope technique wasemployed to produce feedstocks before they were thixoformed at50%liquid fraction.The effect of Mg on the microstructure ofAl-Si-Cu aluminium alloys under as-cast and semisolid conditions was investigated.It was found that by adding Mg to Al-Si-Cualloy,some of the Al2Cu phase and silicon were consumed to form Al5Cu2Mg3Si5and Mg2Si phases.The needle-likeβ-Al5FeSi phasetransformed to Chinese-script-likeπ-Al8Mg3FeSi6with the addition of Mg.In the as-cast alloys,the primaryα(Al)was dendritic,butas the Mg content increased,the phase became less dendritic.Moreover,the Mg addition considerably modified the size of theα(Al)phase,but it had no significant effect on the silicon morphology.In the thixoformed alloys,the microstructure showed a fine globularprimary phase surrounded by uniformly distributed silicon and fragmented intermetallic phases.The eutectic silicon was modifiedfrom a flaky and acicular shape to fine fibrous particles.The effect of Mg on eutectic silicon during semisolid processing wasevident.The primary Mg2Si particles were modified from big polygonal particles to become smaller and more globular,whereas themorphology of the Chinese-script-likeπ-Al8Mg3FeSi6changed to a compact shape.The results also exhibit that as the Mg content inthe A319alloy increased,the hardness,yield strength and ultimate tensile strength of the thixoformed alloys significantly improved,but the elongation to fracture dropped.

ADCP application for long-term monitoring of coastal water

Three kind of application of ADCP is reported for long-term monitoring in coastal sea.(1)The routine monitoring of water qualities. The water quality and ADCP echo data (600 kHz) observed in the long-term are analgzed at MT (Marine Tower) Station of Kansai International Airport in the Osaka Bay, Japan. The correlation between the turbidity and echo intensity in the surface layer is not good because air bubbles generated by breaking wave are not detected by the turbidity meter, but detected well by ADCP. When estimating the turbidity consists of plankton population from echo intensity, the effect of bubbles have to be eliminated. (2) Monitoring stirring up of bottom sediment. The special observation was carried out by using following two ADCP in the Osaka Bay, One ADCP was installed upward on the sea. The other ADCP was hanged downward at the gate type stand about 3 m above from the bottom. At the spring tide, high echo intensities indicating the stirring up of bottom sediment were observed. (3) The monitoring for the boundary condition of water mixing at an estuary. In summer season, the ADCP was set at the mouth of Tanabe Bay in Wakayama Prefecture, Japan. During the observation, water temperature near the bottom showed remarkable falls with interval of about 5-7 d. When the bottom temperature fell, the inflow current with low echo intensity water appears at the bottom layer in the ADCP record. It is concluded that when occasional weak northeast wind makes weak coastal upwelling at the mouth of the bay, the combination of upwelling with internal tidal flow causes remarkable water exchange and dispels the red tide.

Role of Ag addition on microstructure,mechanical properties,corrosion behavior and biocompatibility of porous Ti-30 at%Ta shape memory alloys

In the present study,the thermal,mechanical,and biological properties of xAg/Ti-30Ta(x=0,0.41,0.82 and 2.48 at%)shape memory alloys(SMAs)were investigated.The study was conducted using optical and scanning electron microscopy(SEM),X-ray diffractometry(XRD),compression test,and shape memory testing.The xAg/Ti-Ta was made using a powder metallurgy technique and microwave-sintering process.The results revealed that the addition of Ag has a significant effect on the pore size and shape,whereas the smallest pore size of 11μm was found with the addition of 0.41 at%along with a relative density of 72%.The fracture stress and strain increased with the addition of Ag,reaching the minimum values around 0.41 at%Ag.Therefore,this composition showed the maximum stress and strain at fracture region.Moreover,0.82 Ag/Ti-Ta shows more excellent corrosion resistance and biocompatibility than other percentages,obtaining almost the same behaviour of the pure Ti and Ti-6Al-4V alloys,which can be recommended for their promising and potential response for biomaterial applications.

Conceptual design of oil palm fibre reinforced polymer hybrid composite automotive crash box using integrated approach

A hybrid conceptual design approach was introduced in this study to develop a conceptual design of oil palm polymer composite automotive crash box(ACB). A combination of theory of inventive problem solving(TRIZ), morphological charts and biomimetics was applied where the foremost requirements in terms of the material characteristics, function specifications, force identification, root cause analysis, geometry profile and design selection criteria were considered. The strategy was to use creations of nature to inspire five innovative conceptual designs of the ACB structure and the AHP method was applied to perform the pairwise analysis of selecting the best ACB conceptual design. A new conceptual design for a composite ACB was conceived bearing in mind the properties of natural fibre, unlike those of conventional materials such as steel alloys and aluminium alloys. The design with the highest ranking(26.6 %) was chosen as the final conceptual design, which was the one with a honeycomb structure for the outermost profile, reinforced with a spider web structure inside the part, supported by fibre foam structure extracted from the woodpecker sponge tissue at the centre to maximize the energy absorption capability. The new design could solve the problem of bending collapse which is a major cause of failure to absorb maximum impact energy for ACB during collision. However, the final conceptual design will still need several modifications for production and assembly purposes, which will be completed in a further study.

Microstructure and superelastic properties of free forged Ti-Ni shape-memory alloy

Elemental titanium(Ti)and nickel(Ni)powders were consolidated by spark plasma sintering(SPS)to fabricate Ti-51%Ni(mole fraction)shape-memory alloys(SMAs).The objective of this study is to enhance the superelasticity of SPS produced Ti-Ni alloy using free forging as a secondary process.Products from two processes(with and without free forging)were compared in terms of microstructure,transformation temperature and superelasticity.The results showed that,free forging effectively improved the tensile and shape-memory properties.Ductility increased from 6.8%to 9.2%after forging.The maximum strain during superelasticity increased from 5%to 7.5%and the strain recovery rate increased from 72%to 92%.The microstructure of produced Ti-51%Ni SMA consists of the cubic austenite(B2)matrix,monoclinic martensite(B19′),secondary phases(Ti3Ni4,Ti2Ni and TiNi3)and oxides(Ti4Ni2O and Ti3O5).There was a shift towards higher temperatures in the martensitic transformation of free forged specimen(aged at 500°C)due to the decrease in Ni content of B2 matrix.This is related to the presence of Ti3Ni4 precipitates,which were observed using transmission electron microscope(TEM).In conclusion,free forging could improve superelasticity and mechanical properties of Ti-51%Ni SMA.

Inversion of the three-dimensional temperature structure of mesoscale eddies in the Northwest Pacific based on deep learning

Mesoscale eddies,which are mainly caused by baroclinic effects in the ocean,are common oceanic phenomena in the Northwest Pacific Ocean and play very important roles in ocean circulation,ocean dynamics and material energy transport.The temperature structure of mesoscale eddies will lead to variations in oceanic baroclinity,which can be reflected in the sea level anomaly(SLA).Deep learning can automatically extract different features of data at multiple levels without human intervention,and find the hidden relations of data.Therefore,combining satellite SLA data with deep learning is a good way to invert the temperature structure inside eddies.This paper proposes a deep learning algorithm,eddy convolution neural network(ECN),which can train the relationship between mesoscale eddy temperature anomalies and sea level anomalies(SLAs),relying on the powerful feature extraction and learning abilities of convolutional neural networks.After obtaining the temperature structure model through ECN,according to climatic temperature data,the temperature structure of mesoscale eddies in the Northwest Pacific is retrieved with a spatial resolution of 0.25°at depths of 0–1000 m.The overall accuracy of the ECN temperature structure is verified using Argo profiles at the locations of cyclonic and anticyclonic eddies during 2015–2016.Taking 10%error as the acceptable threshold of accuracy,89.64%and 87.25%of the cyclonic and anticyclonic eddy temperature structures obtained by ECN met the threshold,respectively.

In vitro microbiologically-induced concrete corrosion behavior of Ag^(+) loaded zeolite-polyurethane coating for concrete sewer applications

Microbiologically-induced concrete corrosion(MICC)refers to chemical reactions between biologically produced sulphuric acid and with hydration products in the hardened concrete paste,resulting in an early reduction of strength,deterioration,and very severe circumstances,structural failure.This paper explores the bactericidal characteristics of cementitious materials with surface coated with modified zeolite-polyurethane.The zeolite-polyurethane coating incorporated with silver was studied in environments inoculated with A.thiooxidans bacteria for 8 consecutive weeks.The antibacterial characteristics were evaluated in terms of pH,optical density(OD),sulphate production and bacteria count to determine the effectiveness of the zeolite-polyurethane coatings in environments inoculated with A.thiooxidans bacteria producing the sulphuric acid.The results revealed that the samples incorporated with silver modified zeolites generally showed antibacterial performance(regardless of the zeolite type)compared with unmodified polyurethane coating.This was evaluated by the lack of bacteria attachment and the deformed microcolonies on the sample surface,lag in pH reduction,increase in OD,and sulphate production.The silver zeolites gained their antibacterial performance from the release of silver ions(Ag^(+))when the sample comes into contact with aqueous solutions.This results in the inhibition of cell functions of the bacteria and subsequently causes cell damage.

Microwave sintering effects on the microstructure and mechanical properties of Ti–51at%Ni shape memory alloys

Ti–51at%Ni shape memory alloys(SMAs) were successfully produced via a powder metallurgy and microwave sintering technique.The influence of sintering parameters on porosity reduction,microstructure,phase transformation temperatures,and mechanical properties were investigated by optical microscopy,field-emission scanning electron microscopy(FE-SEM),X-ray diffraction(XRD),differential scanning calorimetry(DSC),compression tests,and microhardness tests.Varying the microwave temperature and holding time was found to strongly affect the density of porosity,presence of precipitates,transformation temperatures,and mechanical properties.The lowest density and smallest pore size were observed in the Ti–51at%Ni samples sintered at 900°C for 5 min or at 900°C for 30 min.The predominant martensite phases of β2 and β19′ were observed in the microstructure of Ti–51at%Ni,and their existence varied in accordance with the sintering temperature and the holding time.In the DSC thermograms,multi-transformation peaks were observed during heating,whereas a single peak was observed during cooling;these peaks correspond to the presence of the β2,R,and β19′ phases.The maximum strength and strain among the Ti–51at%Ni SMAs were 1376 MPa and 29%,respectively,for the sample sintered at 900°C for 30 min because of this sample's minimal porosity.

Manufacturing Process Selection of “Green” Oil Palm Natural Fiber Reinforced Polyurethane Composites Using Hybrid TEA Criteria Requirement and AHP Method for Automotive Crash Box

In this study,the best manufacturing process will be selected to build an automotive crash box using green oil palm natural fibre-reinforced polyurethane composite materials.This paper introduces an approach consist of technical aspects(T),the economic point of view(E)and availability(A),and it’s also called as TEA requirement.This approach was developed with the goal of assisting the design engineer in the selection of the best manufacturing process during the design phase at the criteria selection stage.In this study,the TEA requirement will integrate with the analytical hierarchy process(AHP)to assist decision makers or manufacturing engineers in determining the most appropriate manufacturing process to be employed in the manufacture of a composite automotive crash box(ACB)at the early stage of the product development process.It is obvious that a major challenge in the manufacturing selection process is lack of information regarding manufacturing of ACB using natural fibre composite(NFC).There have been no previous studies that examined ranking manufacturability processes in terms of their suitability.Therefore,the TEA-AHP hybrid method was introduced to provide unprejudiced criteria-ranking selection prior to evaluation of pairwise comparisons.At the end of this study,the pulforming process was selected as the best manufacturing process for fabrication of the ACB structural component.

Assessment of theoretical approaches to derivation of internal solitary wave parameters from multi-satellite images near the Dongsha Atoll of the South China Sea

This study assesses the accuracy and the applicability of the Korteweg-de Vries(KdV)and the nonlinear Schr?dinger(NLS)equation solutions to derivation of dynamic parameters of internal solitary waves(ISWs)from satellite images.Visible band images taken by five satellite sensors with spatial resolutions from 5 m to 250 m near the Dongsha Atoll of the northern South China Sea(NSCS)are used as a baseline.From the baseline,the amplitudes of ISWs occurring from July 10 to 13,2017 are estimated by the two approaches and compared with concurrent mooring observations for assessments.Using the ratio of the dimensionless dispersive parameter to the square of dimensionless nonlinear parameter as a criterion,the best appliable ranges of the two approaches are clearly separated.The statistics of total 18 cases indicate that in each 50%of cases,the KdV and the NLS approaches give more accurate estimates of ISW amplitudes.It is found that the relative errors of ISW amplitudes derived from two theoretical approaches are closely associated with the logarithmic bottom slopes.This may be attributed to the nonlinear growth of ISW amplitudes as propagating along a shoaling thermocline or topography.The test results using three consecutive satellite images to retrieve the ISW propagation speeds indicate that the use of multiple satellite images(>2)may improve the accuracy of retrieved phase speeds.Meanwhile,repeated multi-satellite images of ISWs can help to determine the types of ISWs if mooring data are available nearby.

Influence of B_2O_3 on Spectroscopic Properties of Er^(3+)/Yb^(3+) Co-Doped Tungsten-Tellurite Glasses

A series of novel Er^3＋/Yb^3＋ co-doped （85- x ） TeO2-15WO3-xB2O3 （TWB;x=2%,5%,8%（mole fraction） ） glasses were prepared. Influence of B203 on the spectroscopic properties of Er^3＋/Yb^3＋ co-doped tungsten-tellurite glasses were investigated. It is found that the intensity of 1.5μm fluorescence, lifetime of the ^4I13/2 level and upconversion fluorescence all decrease with the increase of B2O3 content. The product of full width at half maximum （FWHM） and stimulated emission cross-section （σe^peak） of Er^3＋ ：^4I13/2→^4I15/2 transition has an optimum when B203 is 5% （mole fraction）. The emission spectra of Er^3＋ ： ^4I13/2→^4I15/2 transition was analyzed using peak-fit routine, and an equivalent four-level system was proposed to estimate the stark splitting for the 411512 and ^4I13/2 levels of Er^3＋ ions in TWB glasses at room temperature.

A Nine-modes Truncation of the Plane Incompressible Navier-Stokes Equations

In this paper a nine-modes truncation of Navier-Stokes equations for a two-dimensional incompressible fluid on a torus is obtained. The stationary solutions, the existence of attractor and the global stability of the equations are firmly proved. What is more, that the force f acts on the mode ks and k7 respectively produces two systems, which lead to a much richer and varied phenomenon. Numerical simulation is given at last, which shows a stochastic behavior approached through an involved sequence of bifurcations.

An Ensemble Application of Conflict-Resolving ART-Based Neural Networks to Fault Detection and Diagnosis

Accurate fault detection and diagnosis is important for secure and profitable operation of modern power systems.In this paper,an ensemble of conflict-resolving Fuzzy ARTMAP classifiers,known as Probabilistic Multiple Fuzzy ARTMAP with Dynamic Decay Adjustment(PMFAMDDA),for accurate discrimination between normal and faulty operating conditions of a Circulating Water(CW)system in a power generation plant is proposed.The decisions of PMFAMDDA are reached through a probabilistic plurality voting strategy that is in agreement with the Bayesian theorem.The results of the proposed PMFAMDDA model are compared with those from an ensemble of Probabilistic Multiple Fuzzy ARTMAP(PMFAM)classifiers.The outcomes reveal that PMFAMDDA,in general,outperforms PMFAM in discriminating operating conditions of the CW system.