Application of robust machine learning methods to modeling hydrogen solubility in hydrocarbon fuels

Having accurate information about the hydrogen solubility in hydrocarbon fuels and feedstocks is very important in petroleum refineries and coal processing plants. In the present work, extreme gradient boosting (XGBoost), multi-layer perceptron (MLP) trained with Levenberg–Marquardt (LM) algorithm, adaptive boosting support vector regression (AdaBoost?SVR), and a memory-efficient gradient boosting tree system on adaptive compact distributions (LiteMORT) as four novel machine learning methods were used for estimating the hydrogen solubility in hydrocarbon fuels. To achieve this goal, a database containing 445 experimental data of hydrogen solubilities in 17 various hydrocarbon fuels/feedstocks was collected in wide-spread ranges of operating pressures and temperatures. These hydrocarbon fuels include petroleum fractions, refinery products, coal liquids, bitumen, and shale oil. Input parameters of the models are temperature and pressure along with density at 20 °C, molecular weight, and weight percentage of carbon (C) and hydrogen (H) of hydrocarbon fuels. XGBoost showed the highest accuracy compared to the other models with an overall mean absolute percent relative error of 1.41% and coefficient of determination (R²) of 0.9998. Also, seven equations of state (EOSs) were used to predict hydrogen solubilities in hydrocarbon fuels. The 2- and 3-parameter Soave-Redlich-Kwong EOS rendered the best estimates for hydrogen solubilities among the EOSs. Moreover, sensitivity analysis indicated that pressure owns the highest influence on hydrogen solubilities in hydrocarbon fuels and then temperature and hydrogen weight percent of the hydrocarbon fuels are ranked, respectively. Finally, Leverage approach results exhibited that the XGBoost model could be well trusted to estimate the hydrogen solubility in hydrocarbon fuels.     

Deterministic and probabilistic ship pitch prediction using a multi-predictor integration model based on hybrid data preprocessing,reinforcement learning and improved QRNN

The deterministic and probabilistic prediction of ship motion is important for safe navigation and stable real-time operational control of ships at sea. However, the volatility and randomness of ship motion, the non-adaptive nature of single predictors and the poor coverage of quantile regression pose serious challenges to uncertainty prediction, making research in this field limited. In this paper, a multi-predictor integration model based on hybrid data preprocessing, reinforcement learning and improved quantile regression neural network (QRNN) is proposed to explore the deterministic and probabilistic prediction of ship pitch motion. To validate the performance of the proposed multi-predictor integrated prediction model, an experimental study is conducted with three sets of actual ship longitudinal motions during sea trials in the South China Sea. The experimental results indicate that the root mean square errors (RMSEs) of the proposed model of deterministic prediction are 0.0254°, 0.0359°, and 0.0188°, respectively. Taking series #2 as an example, the prediction interval coverage probabilities (PICPs) of the proposed model of probability predictions at 90%, 95%, and 99% confidence levels (CLs) are 0.9400, 0.9800, and 1.0000, respectively. This study signifies that the proposed model can provide trusted deterministic predictions and can effectively quantify the uncertainty of ship pitch motion, which has the potential to provide practical support for ship early warning systems.     

Ecological Planting Imperative: Functional Systems,Not Stylized Ecologies

The ability of landscape architectural projects to mitigate the worst effects of climate change will depend upon designed ecological systems. These systems will be built with plants. Despite the recognition of ecology as an essential driver of landscapes, the professionals of landscape architecture too often lack the knowledge and practical skills to create robust vegetative systems. New approaches and tools are required. This article outlines principles and methods for designing biodiverse plant systems for urban sites. Planting methods that increase species richness, functional diversity, and spatial complexity are emphasized as a way of developing more resilient plantings. Selecting species with similar evolutionary adaptions to stress, disturbance, and competition—as well as creating multi-layered compositions of diverse plant morphologies—allows designers to create compatible, long-lived plant mixes. To balance the increased visual complexity of diverse plant mixes, the article explores design techniques to make plantings more appealing to the public. The strategies explored here are based on the projects, experience, and research of Phyto Studio, a Washington, D.C. based studio. The methods build on work described in the author’s book, Planting in a Post-Wild World, an exploration of how to create designed plant communities.     

Control of self-assembly in micro- and nano-scale systems

Control of self-assembling systems at the micro- and nano-scale provides new opportunities for the engineering of novel materials in a bottom-up fashion. These systems have several challenges associated with control including high-dimensional and stochastic nonlinear dynamics, limited sensors for real-time measurements, limited actuation for control, and kinetic trapping of the system in undesirable configurations. Three main strategies for addressing these challenges are described, which include particle design (active self-assembly), open-loop control, and closed-loop (feedback) control. The strategies are illustrated using a variety of examples such as the design of patchy and Janus particles, the toggling of magnetic fields to induce the crystallization of paramagnetic colloids, and high-throughput crystallization of organic compounds in nanoliter droplets. An outlook of the future research directions and the necessary technological advancements for control of micro- and nano-scale self-assembly is provided.     

Thermodynamic analysis of natural gas reciprocating compressors based on real and ideal gas models

The accurate modelling and investigating effects of various parameters of the reciprocating compressors are important subjects. In this work, based on first law of thermodynamics, conversation of mass and real and ideal gas assumptions, a theoretical analysis has been constructed to simulate natural gas reciprocating compressors. For computing the thermodynamic properties of natural gas based on real gas model, the AGA8 equation of state has been used. Numerical results validated with previous measured values and showed a good agreement. The effects of important parameters such as: angular speed, clearance and pressure ratio have been studied on the performance of the compressors. The results reveal the in-control volume temperature for ideal gas is more than real gas model but the mass flow rate and work for real gas is higher than ideal gas model. On the other hand, the indicated work that required for compression is greater for ideal gas model.     

Thermodynamic models for H2O–CO2–H2 mixtures in near-critical and supercritical regions of water

The PVTx properties of the H₂O–CO₂–H₂ mixtures have significant applications in the technology of supercritical water gasification of coal. Here, we first carry out the molecular dynamics simulations of the PVTx properties of the H₂O–CO₂–H₂ mixtures in the near-critical and supercritical regions of water to generate 600 datasets at 750–1150 K and 4.0–443.5 MPa. The molar fraction of each composition in the ternary mixtures ranges from 10% to 80%. Later we investigate the applicability of a well-known thermodynamic model for the ternary mixtures, namely the Duan-Møller-Weare equation of state (DMW EOS). It is observed that the DMW EOS shows great potential in the prediction of the PVTx properties of the ternary mixtures. However, it is noted that the mixing parameters describing the binary interactions of H₂O–H₂ and CO₂–H₂ are still unknown in the DMW EOS. By determining the missing mixing parameters using the Levenberg-Marquardt algorithm, the accuracy of the original DMW EOS is improved for the ternary mixtures. Moreover, optimizing the coefficients in the DMW EOS further promotes the accuracy of the model for the H₂O–CO₂–H₂ mixtures. The results from this work may facilitate the development of supercritical water gasification of coal.     

Utilizing sensor data to model students’ creativity in a digital environment

While creativity is essential for developing students’ broad expertise in Science, Technology, Engineering, and Math (STEM) fields, many students struggle with various aspects of being creative. Digital technologies have the unique opportunity to support the creative process by (1) recognizing elements of students’ creativity, such as when creativity is lacking (modeling step), and (2) providing tailored scaffolding based on that information (intervention step). However, to date little work exists on either of these aspects. Here, we focus on the modeling step. Specifically, we explore the utility of various sensing devices, including an eye tracker, a skin conductance bracelet, and an EEG sensor, for modeling creativity during an educational activity, namely geometry proof generation. We found reliable differences in sensor features characterizing low vs. high creativity students. We then applied machine learning to build classifiers that achieved good accuracy in distinguishing these two student groups, providing evidence that sensor features are valuable for modeling creativity.     

Integrated smart grid systems security threat model

The smart grid (SG) integrates the power grid and the Information and Communication Technology (ICT) with the aim of achieving more reliable and safe power transmission and distribution to the customers. Integrating the power grid with the ICT exposes the SG to systems security threats and vulnerabilities that could be compromised by malicious users and attackers. This paper presents a SG systems threats analysis and integrated SG Systems Security Threat Model (SSTM). The reference architecture of the SG, with its components and communication interfaces used to exchange the energy-related information, is integrated with the results of SG systems security threat analysis to produce a comprehensive, integrated SG SSTM. The SG SSTM in this paper helps better depict and understand the vulnerabilities exploited by attackers to compromise the components and communication links of the SG. The SG SSTM provides a reference of the systems security threats for industrial security practitioners, and can be used for design and implementation of SG systems security controls and countermeasures.     

Customer satisfaction and loyalty analysis with classification algorithms and Structural Equation Modeling

Businesses can maintain their effectiveness as long as they have satisfied and loyal customers. Customer relationship management provides significant advantages for companies especially in gaining competitiveness. In order to reach these objectives primarily companies need to identify and analyze their customers. In this respect, effective communication and commitment to customers and changing market conditions is of great importance to increase the level of satisfaction and loyalty. To evaluate this situation, level of customer satisfaction and loyalty should be measured correctly with a comprehensive approach. In this study, customers are investigated in 4 main groups according to their level of satisfaction and loyalty with a criteria and group based analysis with a new method. We use classification algorithms in WEKA programming software and Structural Equation Modeling (SEM) with LISREL tools together to analyze the effect of each satisfaction and loyalty criteria in a satisfaction–loyalty matrix and extend the customer satisfaction and loyalty post-analysis research bridging the gap in this field of research. To convert developed conceptual thought to experimental study, white goods industry is exemplified. 15 criteria are used for evaluation in 4 customer groups and a satisfaction–loyalty survey developed by experts is applied to 200 customers with face-to-face interviews. As a result of the study, a customer and criteria grouping method is created with high performance classification methods and good fit structural models. In addition, results are evaluated for developing a customer strategy improvement tool considering method outcomes.     

Factors that determine residents’ acceptance of smart city technologies

ABSTRACT

While some cities attempt to determine their residents’ demand for smart-city technologies, others simply move forward with smart-related strategies and projects. This study is among the first to empirically determine which factors most affect residents’ and public servants’ intention to use smart-city services. A Smart Cities Stakeholders Adoption Model (SSA), based on Unified Theory of Acceptance and Use of Technology (UTAUT2), is developed and tested on a mid-size U.S. city as a case study. A questionnaire was administered in order to determine the influence of seven factors – effort expectancy, self-efficacy, perceived privacy, perceived security, trust in technology, price value and trust in government – on behaviour intention, specifically the decision to adopt smart-city technologies. Results show that each of these factors significantly influenced citizen intention to use smart-city services. They also reveal perceived security and perceived privacy to be strong determinants of trust in technology, and price value a determinant of trust in government. In turn, both types of trust are shown to increase user intention to both adopt and use smart-city services. These findings offer city officials an approach to gauging residential intention to use smart-city services, as well as identify those factors critical to developing a successful smart-city strategy.