Influence of Humidity State on Dynamic Resilient Modulus of Subgrade Soils: Considering Repeated Wetting-Drying Cycles

The service performance of subgrade depends on the dynamic resilient modulus (MR) of subgrade soils. Meanwhile, due to complicated conditions such as rainfall infiltration, high temperature evaporation, and groundwater level fluctuations, it can be safely said that the humidity state and repeated wetting-drying (WD) cycles affect the MR of subgrade soils. The object of this study is to conduct a series of dynamic triaxial tests after WD cycles to investigate the characteristics of the MR under various factors. The main results are as follows: (i) the MR decreased with the increase of deviator stress and rose with the growth of confining pressure; (ii) the humidification effect caused by the increase in moisture content attenuated the MR; (iii) the accumulation of WD cycles damaged the MR; however the decline rate was gradually retarded until it was stable with WD cycles 5 times; (iv) the satisfactory prediction model for the MR of subgrade soils considering WD cycles was proposed and verified. It is expected that the findings can provide valuable contributions for road engineering.

Use of PNUTGRO model for optimization of sowing date and plant spacing to maximize yield of groundnut (Arachis hypogaea L.)

Lkkj & ihuVxzks ¼ih- ,u- ;w- Vh- th- vkj- vks-½ ekWMy dh izkekf.kdrk fl) djus ds fy, 1987 - 90 ds nkSjku vkuan] xqtjkr esa {ks=h; iz;ksx fd, x, gSaA bl ekWMy dk mi;ksx ew¡xQyh dh QhuksykWth] c<+ksrjh] fodkl vkSj iSnkokj dk iwokZuqeku yxkus ds fy, fd;k x;k gSSA ew¡xQyh ds izfr:fir iq"iu] isfxax] Qyh cuus vkSj Qyh idus dh frfFk;ksa] i.khZ {ks=Qy lwpdkad ¼,y- ,- vkbZ-½ tSoHkkj] 'kSfyax dk izfr’kr rFkk iSnkokj dh rqyuk rhu i)fr;ksa uker% th- ,- ;w- th- 10] th- ,- ;w- th- 2 vkSj vkj- vk-sa - 33 - 1 ls izkIr gq, iszf{kr ekuksa ds lkFk dh xbZ gSA izfr:fir ?kVukØe ls iq"iu ds fy, ,d fnu deh rFkk ik¡p fnu dh c<+r dk] isfxxa ds fy, 2 ls 6 fnuksa dh c<+r] Qyh cuus ds fy, 3 fnu dh deh rFkk 6 fnuksa dh c<+r dk vkSj Qyh idus ds fy, 6 fnu dh deh rFkk 5 fnu rd dh c<+r dk varj ik;k x;k gSA okLrfod ekuksa dh rqyuk esa bl ekWMy ls i.khZ {ks=Qy lwpdkad 91-8 ls 105-8 izfr’kr vkSj 'kSfyax dk izfr’kr 81-5 ls 109-8 ik;k x;k gSA bl ekWMy ls ew¡xQyh dh iSnkokj izsf{kr ekuksa dh rqyuk esa 88-5 ls 112-7 izfr’kr rd ikbZ xbZ gSA bl ekWMy ls izkIr ifj.kkeksa ds vk/kkj ij ij yxkrkj pkj Qlyksa vkSj _rqvksa ds laca/k esa ew¡xQyh dh QhuksYkWkth] c<+ksrjh] fodkl vkSj iSnkokj ds ckjs esa iwokZuqeku larks"ktud ik;k x;k gSA ew¡xQyh dh izsf{kr vkSj izfr:fir iSnkokj ds chp 11 izfr’kr dh ?kVc<+ ikbZ xbZ gS ftlls irk pyrk gS fd ekWMy ds vk/kkj ij fd;k x;k iwokZuqeku larks"ktud gSA ,y- ,- vkbZ- dks NksMdj okLrfod ekuksa vkSj izsf{kr ekuksa esa varj ¼Mh-½ 0-03 vkSj 1-77 ds chp jgk gS ftlls ekWMy ds larks"ktud dk;Z djus dk irk pyrk gSA izfr:i.k v/;;uksa ds ifj.kkeksa ls irk pyrk gS fd tc vf/kd o"kkZ gksus dhs laHkkouk gks rks ew¡xQyh ds chtksa dh lkekU; nwjh rFkk cqokbZ ds lkekU; le; dh vis{kk chtksa dks vf/kd ikl&ikl cksdj rFkk cqokbZ yxHkx ,d lIrkg igys djds ew¡xQyh dh vf/kd iSnkokj izkIr dh tk ldrh gSA Field experiments were conducted at Anand, Gujarat during 1987-90 to validate the PNUTGRO model. The model was used to predict phenology, growth, development and yield of groundnut. The simulated flowering, pegging, pod formation and pod maturity dates, leaf area index (LAI), biomass, shelling % and pod yield of groundnut were compared with the observed values for three cultivars viz., GAUG 10, GAUG 2 and Ro-33-1. The simulated phenological events showed a deviation of –1 to +5 days for flowering, +2 to +6 days for peg formation, -3 to +6 days for pod formation and –6 to +5 days for pod maturity of the crop. The model estimated leaf area index within 91.8 to 105.8% and shelling percentage within 81.5 to 109.8% of the actual values. The model simulated the pod yields within 88.5 to 112.7% of the observed values. The results obtained with the model for the four consecutive crops and seasons revealed satisfactory prediction of phenology, growth, development and yield of groundnut. The percent error between observed and simulated pod yield was 11% which indicated satisfactory prediction by the model. The degree of agreement (d) ranged between 0.03 and 1.77 except for LAI indicating satisfactory performance of the model. Results of simulation studies indicated that when there is a possibility of high rainfall higher pod yield can be achieved by adopting closer spacing and early sowing (one week earlier than normal date of sowing) compared to normal spacing and date of sowing.

Application of Deep Learning in Financial Management Evaluation

The competition among enterprises is becoming increasingly fierce. The research on the financial management evaluation model is helpful for enterprises to find possible risks as soon as possible. This paper constructs the financial management evaluation model based on the deep belief network and applies the analytic hierarchy process to determine the weight of financial management evaluation indicators, which is compared with other classical deep learning evaluation methods, such as KNN, SVM-RBF, and SVM linear. It has achieved an accuracy of more than 81%, showing a satisfactory prediction effect, which is of great significance to formulate corresponding countermeasures, strengthen financial management, improve the capital market system, and promote high-quality economic development. In addition, aiming at the problem of abnormal financial data, this paper uses the new sample dataset obtained by principal component analysis for convolution neural network model learning, which enhances the prediction accuracy of the model and fully shows that deep learning is feasible in the research of financial management prediction, and there is still a lot of space to explore.

VinegarScan: A Computer Tool Based on Ultraviolet Spectroscopy for A Rapid Authentication of Wine Vinegars

Ultraviolet-visible (UV-vis) spectroscopy has shown successful results in the last few years to characterize and classify wine vinegar according to its quality, particularly those with a protected designation of origin (PDO). Due to these promising results, together with the simplicity, price, speed, portability of this technique and its ability to create robust hierarchical classification models, the objective of this work was the development of a computer tool or software, named VinegarScan, which uses the UV-vis spectra to be able to perform quality control and authentication of wine vinegar in a quick and user-friendly way. This software was based on the open-source GUI created in C++ using several data mining algorithms (e.g., decision trees, classification algorithms) on UV-vis spectra. This software achieved satisfactory prediction results with the available analytical UV-vis data. The future idea of utility is to combine the VinegarScan tool with a portable UV-vis device that could be used by control bodies of the wine vinegar industry to achieve a clear differentiation from their competitors to avoid fraud.

A Novel Approach to the Prediction of Biomagnification Factors Based on Molecular Structure Images

Although biomagnification factor (BMF) is an important index of pollutants in food chains, its experimental determination is quite tedious. In this contribution, as the feature information, Tchebichef moments (TMs) were calculated directly from the molecular structural images, and then stepwise regression was employed to establish the prediction model of the logBMF. The proposed approach was applied to the logBMF prediction of organochlorine pollutants, and the correlation coefficient with leave-one-out cross-validation (Rcv) of the obtained model was 0.96, and the root mean square error (RMSEp) for the external independent test set was 0.21. Compared with traditional two-dimensional (2D) quantitative structure-property relationship (QSPR) as well as the reported method, the proposed approach was more simple, accurate and reliable. This study not only obtained the satisfactory prediction model for organochlorine pollutants, but also provided another effective approach to QSPR research.

Evaluation of Machine Learning Models for Estimating PM2.5 Concentrations across Malaysia

Southeast Asia (SEA) is a hotspot region for atmospheric pollution and haze conditions, due to extensive forest, agricultural and peat fires. This study aims to estimate the PM2.5 concentrations across Malaysia using machine-learning (ML) models like Random Forest (RF) and Support Vector Regression (SVR), based on satellite AOD (aerosol optical depth) observations, ground measured air pollutants (NO2, SO2, CO, O3) and meteorological parameters (air temperature, relative humidity, wind speed and direction). The estimated PM2.5 concentrations for a two-year period (2018–2019) are evaluated against measurements performed at 65 air-quality monitoring stations located at urban, industrial, suburban and rural sites. PM2.5 concentrations varied widely between the stations, with higher values (mean of 24.2 ± 21.6 µg m−3) at urban/industrial stations and lower (mean of 21.3 ± 18.4 µg m−3) at suburban/rural sites. Furthermore, pronounced seasonal variability in PM2.5 is recorded across Malaysia, with highest concentrations during the dry season (June–September). Seven models were developed for PM2.5 predictions, i.e., separately for urban/industrial and suburban/rural sites, for the four dominant seasons (dry, wet and two inter-monsoon), and an overall model, which displayed accuracies in the order of R2 = 0.46–0.76. The validation analysis reveals that the RF model (R2 = 0.53–0.76) exhibits slightly better performance than SVR, except for the overall model. This is the first study conducted in Malaysia for PM2.5 estimations at a national scale combining satellite aerosol retrievals with ground-based pollutants, meteorological factors and ML techniques. The satisfactory prediction of PM2.5 concentrations across Malaysia allows a continuous monitoring of the pollution levels at remote areas with absence of measurement networks.

Impact of Radar Reflectivity and Lightning Data Assimilation on the Rainfall Forecast and Predictability of a Summer Convective Thunderstorm in Southern Italy

Heavy and localized summer events are very hard to predict and, at the same time, potentially dangerous for people and properties. This paper focuses on an event occurred on 15 July 2020 in Palermo, the largest city of Sicily, causing about 120 mm of rainfall in 3 h. The aim is to investigate the event predictability and a potential way to improve the precipitation forecast. To reach this aim, lightning (LDA) and radar reflectivity data assimilation (RDA) was applied. LDA was able to trigger deep convection over Palermo, with high precision, whereas the RDA had a key role in the prediction of the amount of rainfall. The simultaneous assimilation of both data sources gave the best results. An alert for a moderate–intense forecast could have been issued one hour and a half before the storm developed over the city, even if predicting only half of the total rainfall. A satisfactory prediction of the amount of rainfall could have been issued at 14:30 UTC, when precipitation was already affecting the city. Although the study is centered on a single event, it highlights the need for rapidly updated forecast cycles with data assimilation at the local scale, for a better prediction of similar events.

Effect of debris accumulation on scour evolution at bridge pier in bank proximity

Large debris transported by flood affects scour features at bridge piers and increases the risks of structural failure. Geometric characteristics of the debris and the relative position of the pier with respect to the river bank are important parameters for the scour process. The interaction between the water flow and debris accumulation increases the shear stress, turbulence and consequently enhances the scour depth at the pier. This paper aims at analyzing such effects on scour evolution at bridge piers. To this end, two series of tests were carried out under clear water condition with different debris geometries and percentage blockage ratios. Experimental evidences showed that the pier position only influences scour evolution and equilibrium morphology for low water depths. Conversely, its effect becomes negligible for scour at bridge piers with debris accumulation and higher water depths. Useful practical relationships have been derived, with satisfactory prediction capability of the scour evolution for all the tested configurations.

Predictive modeling of clinical trial terminations using feature engineering and embedding learning

In this study, we propose to use machine learning to understand terminated clinical trials. Our goal is to answer two fundamental questions: (1) what are common factors/markers associated to terminated clinical trials? and (2) how to accurately predict whether a clinical trial may be terminated or not? The answer to the first question provides effective ways to understand characteristics of terminated trials for stakeholders to better plan their trials; and the answer to the second question can direct estimate the chance of success of a clinical trial in order to minimize costs. By using 311,260 trials to build a testbed with 68,999 samples, we use feature engineering to create 640 features, reflecting clinical trial administration, eligibility, study information, criteria etc. Using feature ranking, a handful of features, such as trial eligibility, trial inclusion/exclusion criteria, sponsor types etc., are found to be related to the clinical trial termination. By using sampling and ensemble learning, we achieve over 67% Balanced Accuracy and over 0.73 AUC (Area Under the Curve) scores to correctly predict clinical trial termination, indicating that machine learning can help achieve satisfactory prediction results for clinical trial study.

Investigation of large-diameter flange joint with soft-gasket

This study focuses on behaviour of the flanged joint with a soft rubber gasket, which is a common type of gasket for evaporator structures. The assembled structure is tested according to EN 13445-3, and strain values are measured using strain gauges. The numerical analysis in ABAQUS comprises several models with different levels of detail. These models are calibrated according to test data. Recommendations regarding the model complexity that is required to obtain satisfactory prediction of non-linear behaviour of flange joints are presented and documented.