Analysis, Predictive Modeling and Hoisted Object Impact Control in Hydro-cylinder Stage- Switching

Solar radiation prediction is an important process in ensuring optimal exploitation of solar energy power. Numerous models have been applied to this problem, such as numerical weather prediction models and artificial intelligence models. However, well-designed hybridization approaches that combine numerical models with artificial intelligence models to yield a more powerful model can provide a significant improvement in prediction accuracy. In this paper, novel hybrid machine learning approaches that exploit auxiliary numerical data are proposed. The proposed hybrid methods invoke different machine learning paradigms, including feature selection, classification, and regression. Additionally, numerical weather prediction (NWP) models are used in the proposed hybrid models. Feature selection is used for feature space dimension reduction to reduce the large number of recorded parameters that affect estimation and prediction processes. The rough set theory is applied for attribute reduction and the dependency degree is used as a fitness function. The effect of the attribute reduction process is investigated using thirty different classification and prediction models in addition to the proposed hybrid model. Then, different machine learning models are constructed based on classification and regression techniques to predict solar radiation. Moreover, other hybrid prediction models are formulated to use the output of the numerical model of Weather Research and Forecasting (WRF) as learning elements in order to improve the prediction accuracy. The proposed methodologies are evaluated using a data set that is collected from different regions in Saudi Arabia. The feature-reduction has achieved higher classification rates up to 8.5% for the best classifiers and up to 15% for other classifiers, for the different data collection regions. Additionally, in the regression, it achieved improvements of average root mean square error up to 5.6% and in mean absolute error values up to 8.3%. The hybrid models could reduce the root mean square errors by 70.2% and 4.3% than the numerical and machine learning models, respectively, when these models are applied to some dataset. For some reduced feature data, the hybrid models could reduce the root mean square errors by 47.3% and 14.4% than the numerical and machine learning models, respectively.

This paper was removed from publication at the author’s request, August 30, 2021. Copyright © 2021 by ASME

Since PM2.5 (particulate matter with an aerodynamic diameter of less than 2.5 µm) directly threatens public health, in order to take appropriate measures(prevention) in advance, the Korea Ministry of Environment(MOE) has been implementing PM10 forecast nationwide since February 2014. PM2.5 forecasts have been implemented nationwide since January 2015. The currently implemented PM forecast by the MOE subdivides the country into 19 regions, and forecasts the level of PM in 4 stages of “Good”, “Moderate”, “Unhealthy”, and “Very unhealthy”.Currently PM air quality forecasting system operated by the MOE is based on a numerical forecast model along with a weather and emission model. Numerical forecasting model has fundamental limitations such as the uncertainty of input data such as emissions and meteorological data, and the numerical model itself. Recently, many studies on predicting PM using artificial intelligence such as DNN, RNN, LSTM, and CNN have been conducted to overcome the limitations of numerical models.In this study, in order to improve the prediction performance of the numerical model, past observational data (air quality and meteorological data) and numerical forecasting model data (chemical transport model) are used as input data. The machine learning model consists of DNN and Seq2Seq, and predicts 3 days (D+0, D+1, D+2) using 6-hour and 1-hour average input data, respectively. The PM2.5 concentrations predicted by the machine learning model and the numerical model were compared with the PM2.5 measurements.The machine learning models were trained for input data from 2015 to 2020, and their PM forecasting performance was tested for 2021. Compared to the numerical model, the machine learning model tended to increase ACC and be similar or lower to FAR and POD.Time series trend was showed machine learning PM forecasting trend is more similar to PM measurements compared with numerical model. Especially, machine learning forecasting model can appropriately predict PM low and high concentrations that numerical model is used to overestimate.Machine learning forecasting model with DNN and Seq2Seq can found improvement of PM forecasting performance compared with numerical forecasting model. However, the machine learning model has limitations that the model can not consider external inflow effects.In order to overcome the drawback, the models should be updated and added some other machine learning module such as CNN with spatial features of PM concentrations.  Acknowledgements This study was supported in part by the ‘Experts Training Graduate Program for Particulate Matter Management’ from the Ministry of Environment, Korea and by a grant from the National Institute of Environmental Research (NIER), funded by the Ministry of Environment (ME) of the Republic of Korea (NIER-2022-04-02-068).  

Tensile fracture of clay soils either due to loading or due to desiccation is a common problem encountered in many geotechnical, geoenvironmental and resources engineering applications such as in compacted clay liners, dams, embankments, slopes, seabed trenching for pipeline placement and in mine tailings. However, the fundamental understanding of this process and its modelling capability has not yet advanced satisfactorily. This research intends to fill this gap, following on the past and concurrent research undertaken on this topic by Monash Geomechanics Group. The current research is to develop fundamental characteristics of fracture properties, develop relevant measuring and analysis techniques and provide the basis for theoretical modelling. The research undertaken comprised of three main laboratory testing stages, and analytical, numerical, theoretical and predictive modelling. Five main different soils were used throughout this thesis including: Werribee clay, Merri Creek clay, Altona North clay, Prestige NY kaolin clay and HR1F kaolin clay. The first three are naturally available in Victoria whereas the two kaolin clays are sourced from commercial dealers in NSW. A comprehensive soil properties database was compiled for all soils tested. Advanced image analysis techniques were extensively used throughout testing to capture strains caused by loading and/or desiccation and determine fracture propagation surfaces. Tensile crack surfaces of compacted soils with varying compaction pressure were analysed on a macro scale to identify voids and aggregate conglomeration. The tensile strength of soils was rigorously tested for mechanical loading and desiccation induced cracking. Mechanically loaded samples were examined for effects of preconsolidation pressure, compaction pressure, soil type and water content at failure. Tensile loading tests were completed using the indirect diametrical tensile (IDT) test. Results on tensile strength found from past literature were compiled and analysed using the MPK framework for volumetric behaviour of unsaturated soils. A line of optimum tensile strength was found from void ratio and moisture ratio for various soil types. An extensive restrained shrinkage desiccation test (Monash desiccation cracking test) was introduced to determine tensile strength, fracture toughness, shrinkage strains and suction from changing water content. Tests were modelled using analytical and numerical modelling. A theoretical and predictive model was determined using MIT and critical state methods based on the restrained shrinkage desiccation test. Fracture properties of clay soils were analysed under four-point bending notch beams and cylindrical ring tests. Linear elastic fracture mechanics, elastic-plastic fracture mechanics and plastic fracture mechanics were all used in calculating fracture energy and toughness. Numerical modelling was undertaken using FLAC3D and UDEC codes to model laboratory and analytical test results for restrained shrinkage tests. UDEC was used to model fracture properties from laboratory restrained tests. Finally, comparisons between different tensile strength tests and numerical models were examined.

As universities around the world have begun to use learning management systems (LMSs), more learning data have become available to gain deeper insights into students' learning processes and make data‐driven decisions to improve student learning. With the availability of rich data extracted from the LMS, researchers have turned much of their attention to learning analytics (LA) applications using educational data mining techniques. Numerous LA models have been proposed to predict student achievement in university courses. To design predictive LA models, researchers often follow a data‐driven approach that prioritizes prediction accuracy while sacrificing theoretical links to learning theory and its pedagogical implications. In this study, we argue that instead of complex variables (e.g., event logs, clickstream data, timestamps of learning activities), data extracted from online formative assessments should be the starting point for building predictive LA models. Using the LMS data from multiple offerings of an asynchronous undergraduate course, we analysed the utility of online formative assessments in predicting students' final course performance. Our findings showed that the features extracted from online formative assessments (e.g., completion, timestamps and scores) served as strong and significant predictors of students' final course performance. Scores from online formative assessments were consistently the strongest predictor of student performance across the three sections of the course. The number of clicks in the LMS and the time difference between first access and due dates of formative assessments were also significant predictors. Overall, our findings emphasize the need for online formative assessments to build predictive LA models informed by theory and learning design. Practitioner notes What is already known about this topic Higher education institutions often use learning analytics for the early identification of low‐performing students or students at risk of dropping out. Most predictive models in learning analytics rely on immutable student characteristics (e.g., gender, race and socioeconomic status) and complex variables extracted from log data within a learning management system. Prioritizing prediction accuracy without theory orientation often yields “black‐box” models that fail to inform educators on what remedies need to be taken to improve student learning. What this paper adds Predictive models in learning analytics should consider learning theory, pedagogy and learning design to identify key predictors of student learning. Online formative assessments can be a starting point for building predictive models that are not only accurate but also provide educators with actionable insights on how student learning can be improved. Time‐related and score‐related features extracted from online formative assessments are particularly useful for predicting students' course performance. Implications for practice and/or policy This study provides strong evidence for using online formative assessments as the foundation for predictive models in learning analytics. Student data from online formative assessments can help educators provide students with feedback while informing future formative assessment cycles. Higher education institutions should avoid the hype around complex data from learning management systems and instead rely on effective learning tools such as online formative assessments to revolutionize the use of learning analytics.

The toxic gas produced by the underground large-scale contained explosions have a great impact on the surrounding environment safety of the explosion project, so it is of great significance for the safety protection design to predict the leakage of the explosion gases to the environment. Based on the study of the one-dimensional spherically (axially) symmetric fluid dynamic model for seepage of the underground contained explosion gases, the qualitative and quantitative variables which impact the gas seepage, such as the explosion cavity, the gas pressure in the cavity and the rock mass permeability, among which the qualitative variables can describe the uncertainty of the geological mass, were analyzed. In accordance with the physical significance of variables and the quantification theory, the qualitative variables were quantified and such details as the principle for variable selection, significance assessment and recurrence test to establish a quantification prediction model, etc. were discussed. Based on 15 test samples on the hard rock conditions, we have established a quantification model for prediction of the contained explosion gases leakage at the specific site. The prediction accuracy of this model can meet the requirement of this project, and according to this model, it is known that such qualitative factors as groundwater status and rock properties have significant influence on the gas leakage. In this study, the prediction model is established based on physical analysis so that the quantitative prediction model could have its rationality and effectiveness well ensured, and the research methods and results in this paper could be promoted and applied in the similar engineering practices.

Fiber waviness is one of the most significant defects that occurs in composites due to the severe knockdown in mechanical properties that it causes. This paper investigates the mechanisms for the generation of fiber path defects during processing of composites prepreg materials and proposes new predictive numerical models. A key focus of the work was on thick sections, where consolidation of the ply stack leads to out of plane ply movement. This deformation can either directly lead to fiber waviness or can cause excess fiber length in a ply that in turn leads to the formation of wrinkles. The novel predictive model, built on extensive characterization of prepregs in small-scale compaction tests, was implemented in the finite element software abaqus as a bespoke user-defined material. A number of industrially relevant case studies were investigated to demonstrate the formation of defects in typical component features. The validated numerical model was used to extend the understanding gained from manufacturing trials to isolate the influence of various material, geometric, and process parameters on defect formation.

Machine learning algorithms, namely artificial neural network modeling, were used to create Prediction models for force and moment coefficients of axisymmetric bodies of revolution. These prediction models had highly nonlinear functional relationships to both geometric parameters and inflow conditions, totaling five input factors. A uniform experimental design was created consisting of 50 design points in these five factors and dictated which test points to simulate. Data was generated using computational fluid dynamic simulations, which were performed on all geometries using NavyFOAM at the experimental conditions prescribed by the designed experiment. The prediction models were validated by comparing behavioral trends in responses to previous research conducted by the author on a similar geometry. A test data sets was also created and used to ensure that the prediction models were not overfit to the training data and that they could accurately predict arbitrary geometries and inflow conditions within the experimental design region. Once the prediction models were validated, they were used to study the effects of varying the geometric parameters, inherent to the experiment, on each of the force and moment coefficients. Introduction Multidisciplinary optimization (MDO) schemes used in the early concept design phases for aero/hydrodynamic vehicles often use simplified planar maneuvering characteristics based on empirical or analytical relations in order to limit the computational cost of maneuverability prediction. This method leaves a more detailed analysis of the maneuvering behavior of a design to later in the process, where improvement or correction of an adverse behavior may be difficult to implement. The analysis of out-of-plane conditions or combined pitch-yaw conditions especially, are usually relegated to the detail analysis phase as empirical/ analytical descriptions of these conditions are lacking in the literature. It is therefore desired to develop a method to move these more detailed maneuvering analyses forward in the design phase.

Being able to accurately predict temperature from the solar power hitting the photovoltaic panels is a key to estimate the various parameters which helps in temperature prediction. The solar power coming to our planet is predictable, but the energy produced fluctuates with varying atmospheric conditions. Usually, numerical weather prediction models are used to make irradiation forecasts. Our study based on back propagation neural network which is trained and tested based on dataset provided. This paper utilizes artificial neural networks for temperature forecasting. Our study based on back propagation neural network which is trained and tested based on dataset provided. In formulating the ANN-based predictive model; three-layer network has been constructed. Suitable air temperature predictions can provide farmers and producers with valuable information when they face decisions regarding the use of mitigating technologies such as orchard heaters or irrigation. Temperature warnings are important forecasts because they are used to protect life and property. Temperature forecasting is the application of science and technology to predict the state of the temperature for a future time and a given location. Temperature forecasts are made by collecting quantitative data about the current state of the atmosphere. In this paper, a neural network-based algorithm for predicting the temperature is presented. The Neural Networks package supports different types of training or learning algorithms. One such algorithm is Back Propagation Neural Network (BPN) technique. The main advantage of the BPN neural network method is that it can fairly approximate a large class of functions. This method is more efficient than numerical differentiation. The simple meaning of this term is that our model has potential to capture the complex relationships between many factors that contribute to certain temperature. The proposed idea is tested using the real time dataset. The results are compared with practical working of meteorological department and these results confirm that our model have the potential for successful application to temperature forecasting. Real time processing of weather data indicate that the BPN based weather forecast have shown improvement not only over guidance forecasts from numerical models, but over official local weather service forecasts as well. Artificial neural networks and the back propagation algorithm used for temperature forecasting in general are explained.

Fully-coupled hydroelastic modeling of a deformable wall in waves

Objective: To meet increasing customer demand, many vehicle manufacturers are now offering a panoramic sunroof option in their vehicle lineup. Currently, there is no regulatory or consumer test aimed at assessing the potential for ejection mitigation of roof glazing, which leaves manufacturers to develop internal performance standards to guide designs. The goal of this study was to characterize the variety of occupant-to-roof impacts involving unbelted occupants in rollover crashes to determine the ranges of possible effective masses and impact velocities. This information can be used to define occupant retention requirements and performance criteria for roof glazing in occupant ejection protection.Methods: This study combined computational (MADYMO and LS-Dyna) simulations of occupant kinematics in rollover crashes with laboratory rollover crash tests using the dynamic rollover test system (DRoTS) and linked them through controlled anthropomorphic test device (ATD)-to-roof (“drop”) impact tests. The DRoTS and the ATD drop tests were performed to explore impact scenarios and estimate dummy-to-roof impact impulses. Next, 13 sets of vehicle kinematics and deformation data were extracted from a combination of vehicle dynamics and finite element model simulations that reconstructed variations of rollover crash cases from the field data. Then occupant kinematics data were extracted from a full-factorial sensitivity study that used MADYMO simulations to investigate how changes in anthropometry and seating position would affect occupant–roof impacts across all 13 cases. Finite element (FE) simulations of ATD and Global Human Body Models Consortium (GHBMC) human body model (HBM) roof impacts were performed to investigate the most severe cases from the MADYMO simulations to generate a distribution of head-to-roof impact energies.Results: From the multiparameter design of experiment and experimental study, kinematics and energy output were extracted and analyzed. Based on dummy-to-roof impact force and dummy-to-roof impact velocity, the most severe rollover scenarios were identified. In the DRoTS experiments followed by the drop tests, the range of identified impact velocities was between 2 and 5.8 m/s. However, computational simulations of the rollover crashes showed higher impact velocities and similar effective masses. The largest dummy-to-roof impact velocity was 11 m/s.Conclusions: This study combined computational and experimental analyses to determine a range of possible unbelted occupant-to-roof impact energies. These results can be used to determine design parameters for an impactor for the assessment of the risk of roof glazing ejection for unbelted occupants in rollover crashes.

The full time response of a space frame under impact loading perpendicular to the frame plane is discussed. Theoretical solutions and experimental results are presented and compared. A space frame clamped at its two ends is loaded by a 0.22 lead bullet that hits a mass in the middle of the transversal beam of the frame. The loading time is about 40 to 60 ?sec and the impact impulses in experiment from 0.5 to 1 Ns. The time response of this frame can be divided into four phases where different physical effects are dominant: (a) the ‘loading’ phase where elastic wave motion dominates the time response. Because of the high impact impulses, plastic deformation occurs in the vicinity of the mass and must be included in a theoretical model. The influence of reflections at the corners on the time response is shown; (b) the ‘evolution’ phase. Within this phase, a plastic collapse mechanism develops. Most of this phase is dominated by elastic deformation but local plastic deformations beside the mass are also present. Because many reflections at the corners and the clampings occur within this phase, a modal analysis method is used to predict time histories; (c) the ‘plastic’ phase with plastic zones at the clampings. This phase sets in after the bending wave reaches the clampings. It is characterized by plastic deformation near the clampings and elastic deformation of the other parts of the frame. We used a modal analysis including plastic ‘modes’ to get accurate results; (d) the ‘elastic vibration’ phase.

The results of laboratory experiments and numerical model simulations are described in which the motion of a round, negatively-buoyant, turbulent jet discharged horizontally above a slope into a rotating homogeneous fluid has been investigated. For the laboratory study, flow visualisation data are presented to show the complex three-dimensional flow fields generated by the discharge. Analysis of the experimental data indicates that the spatial and temporal developments of the flow field are controlled primarily by the lateral and vertical discharge position of the jet (with respect to the bounding surfaces of the container of width W) and the specific momentum (M 0) and buoyancy (B 0) fluxes driving the jet. The flow is seen to be characterised by the formation of (i) a primary anticyclonic eddy (PCC) close to the source, (ii) an associated secondary cyclonic eddy (SCE) and (iii) a buoyancy-driven bottom boundary current along the right side boundary wall. For the parameter ranges studied, the size L p, s and spatial location x p, s of the PCC and SCE (and the nose velocity u N of the boundary current) are shown to be only weakly-dependent upon the value of the mixed parameter M 0Ω/B 0, where Ω is the background rotation rate. Both L p and x p are shown to scale with the separation distance y*/W of the right side wall (y = 0) from the source (y = y*), both L s and x s scale satisfactorily with the length scale l M (= M 0 3/4/B 0 ½) and u N is determined by the appropriate gravity current speed [(g']0 H]½ and the separation distance y*/W. Numerical model results show good qualitative agreement with the laboratory data with regard to the generation of the PCC, SCE and boundary current as characteristic features of the flow in question. In addition, extension of the numerical model to diagnose potential vorticity and plume thickness distributions for the laboratory cases allow the differences in momentum-and buoyancy-dominated flows to be clearly delineated. Specifically, the characteristic features of the SCE are shown to be strongly dependent upon the value of M 0Ω/B 0 for the buoyant jet flow; not least, the numerical model data are able to confirm the controlling role played by the boundary walls in the laboratory experiments. Quantitative agreement between the numerical and laboratory model data is fair; most significantly, the success of the former model in simulating the dominant flow features from the latter enables the reliable extension of the numerical model to be made to cases of direct oceanic interest.

Debris flows are massively erosive mass movements that pose an increasing threat to infrastructure and settlements in mountainous areas due to more intense heavy rainfall events in the future. A major contributor to the magnitude for runoff generated debris flow is the parameter of effective erosion. It directly translates to the hazard potential of debris flows, but it is yet to be sufficiently implemented in models to achieve a predictive performance.We developed a simple predictive erosive debris-flow model calibrated on active channels in the northern Bavarian Alps. The debris-flows at the study sites recently occurred in 2015 and in 2021, and all entrained more than 80% of their final volume from the sediment channel bed. Geomorphic change was calculated from pre- and post-event LiDAR data and the total volume of the flow was then compared to catchment characteristics. For a detailed analysis we divided the channel into equal segments and compared the respective eroded volume to flow parameters of the adjacent cross section which were simulated in a numerical model. The initiation volume was estimated by a runoff calculation from the respective heavy precipitation events recorded with radar data. We were able to obtain a correlation that can be used in a predictive debris-flow model to iteratively calculate the erosion for runoff-generated debris flows that are triggered by intense rainstorms. This model allows improved predictions of the magnitude of debris-flow prone channels through a forward-modelling approach.

Aseismic ability estimation of school building using predictive data mining models