Coal use needs to rapidly decline in the global energy mix in the next few decades in order to meet the Paris climate goals of keeping global warming well below 2-degrees Celsius. In emerging economies such as India (the second largest producer and consumer of coal) this would entail reducing long-term coal dependency. Prior work has focused on a coal transition in India from a techno-economic point of view, yet little attention has been given to the socio-economic dimensions of this transition. This is in part due to lack of availability of datasets required for such analysis. The first step in understanding the socio-economic dimensions of a coal transition in India is to understand the scale of current socio-economic dependency on coal at the sub-national level. We contribute to this literature by creating a novel dataset comprised of all 459 operational coal mines in India, using multiple Right to Information Act applications (India's Freedom of Information Act) and then combining this dataset with coal company wise employment factors to estimate direct job numbers at the district level (a sub-administrative unit). We find that coal is produced in 51 districts in 13 states in India with large variations in employment numbers among these districts. While Korba district in Chhattisgarh state is the highest coal producing district, Dhanbad district in Jharkhand state is home to the highest number of coal mining workers. This is the first attempt at understanding the socio-economic dependency on coal at a district level and future work could focus on quantifying other district level socio-economic indicators such as coal related revenues. The new dataset and the results of this paper will be useful for scholars conducting future work on coal transitions and related topics.

This article presents a detail experimental procedure to perform float-and-sink tests for classifying coal samples according to their densities. Moreover, this article emphasizes obtaining 'partition curves' for three different coal samples (heavy media bath, big barrel and small barrel), which helps in evaluating and demonstrating classifier performance. Calculations of independent variables such as 'Probable Error' and 'Imperfection' are also discussed for partition curve that helps in evaluating the effectiveness of various beneficiation equipment used for the upgradation of quality of coal received from different coal mines. It was observed that there is a tendency for the partition curves to steepen as the density of separation decreases. In other words, separations at lower density is sharper than separations at higher density.

In this short paper, the hyperbola-based analysis of wave interference recently developed by Thomas is employed to study the pattern of concentric circular fringes captured on a distant screen, when its plane is oriented orthogonal to the line joining two coherent point-sources. Newton's rings are another instance of concentric circular fringes, generated with an extended source and a plano-convex lens-glass plate combination. The underlying geometry of the latter scenario, as described in the literature, allows for the estimation of two important physical parameters viz. the wavelength of light and the refractive index of a liquid medium. It is demonstrated here that the geometrical arrangement of the former scenario can in principle, be utilized to reach the same ends as well. Additionally, the hallmark distinguishing features of both types of circular fringe patterns are qualitatively and quantitatively elucidated.

We propose a simple algorithm to locate the 'corner' of an L-curve, a function often used to select the regularisation parameter for the solution of ill-posed inverse problems. The algorithm involves the Menger curvature of a circumcircle and the golden section search method. It efficiently finds the regularisation parameter value corresponding to the maximum positive curvature region of the L-curve. The algorithm is applied to some commonly available test problems and compared to the typical way of locating the l-curve corner by means of its analytical curvature. The application of the algorithm to the data processing of an electrical resistance tomography experiment on thin conductive films is also reported.

In this note, we contribute a multi-language handwritten digit recognition dataset named MNIST-MIX, which is the largest dataset of the same type in terms of both languages and data samples. With the same data format with MNIST, MNIST-MIX can be seamlessly applied in existing studies for handwritten digit recognition. By introducing digits from 10 different languages, MNIST-MIX becomes a more challenging dataset and its imbalanced classification requires a better design of models. We also present the results of applying a LeNet model which is pre-trained on MNIST as the baseline.

Vulnerability detection and safety of smart contracts are of paramount importance because of their immutable nature. Symbolic tools like OYENTE and MAIAN are typically used for vulnerability prediction in smart contracts. As these tools are computationally expensive, they are typically used to detect vulnerabilities until some predefined invocation depth. These tools require more search time as the invocation depth increases. Since the use of smart contracts increases rapidly, their analysis becomes difficult using these traditional tools. Recently, a machine learning technique called Long Short Term Memory (LSTM) has been used to predict the vulnerability of a smart contract. In the present article, we present how to classify smart contracts into Suicidal, Prodigal, Greedy, or Normal categories using Average Stochastic Gradient Descent Weight-Dropped LSTM (AWD-LSTM), a variant of LSTM. We reduced the class imbalance by considering only distinct opcode combinations for normal contracts and achieved a weighted average F1 score of 90.0%. Such techniques can be utilized in real-time to analyze a large number of smart contracts and to improve their security.

A technique based on Archimedes' Principle is described for measuring the volume of small objects (0.5–5 cm³) less dense than water. The volume of 10 small red chillies was measured by pycnometry and an immersion Archimedes technique which involved suspending the chillies in water in a container placed on an electronic balance. A pycnometer, which uses helium gas is able to determine the internal solid volume of the chillies and water immersion the outer volume. The difference between the two volumes gives the volume of internal air cavities in the chillies. The pycnometer and immersion techniques were compared by measuring the volume of wax candles with volumes between 0.3 and 2.5 cm³. A Bland-Altman analysis revealed that the Archimedes volumes were lower than pycnometry volumes by $6.1\pm 2.3 \%$ which needs further investigation. A combination of pycnometry and water immersion may be a useful tool for botanical studies.

Internal combustion engine vehicles have long dominated personal vehicle sales in the United States. As electric vehicles gain market share, understanding fuel costs of gasoline versus electric vehicles (EVs) is increasingly relevant for potential buyers and for researchers. In particular, understanding the influence of fuel price volatility on fuel costs is of interest. Although it is well known that fueling an EV is cheaper than fueling a gasoline vehicle, costs are less commonly expressed in explicit spatially and temporally resolved terms. This note presents an Excel model that includes state-specific gasoline and residential electricity price data for US states between 2015–2020, including both average overall electricity prices and lowest-cost time-of-use electricity rates (measured over a consecutive 8-hour potential residential charging period) for each state. This model, which we call the Fuel Cost Calculator, also allows users to evaluate the impact of several major external and driver behavior factors on fuel costs for gasoline versus electric vehicles. These combinations of these factors result in unique scenarios which make price differences for the fuel costs more variable. With the Fuel Cost Calculator, the user can create specific scenarios reflecting potential real-world conditions to create cost estimates for gasoline and residential EV fueling over user-determined time periods and distances. Fuel costs for gasoline vehicles are generally higher than for electric vehicles in the US, consistent with prior studies. We also explicitly show that fuel cost budgets are much more volatile for gasoline versus electric vehicles charged at a residence, given the much less stable price of oil versus electricity. This volatility is sufficiently large so as to be relevant for vehicle owner budgets. The FCC also shows carbon dioxide (CO₂) emissions associated with vehicle use, resolved at the state level.

Purpose: This study suggested a new method predicting the dose-volume parameter for radiation treatment planning evaluation using machine learning, and to evaluate the performance of different learning algorithms in the parameter prediction. Methods: Dose distribution index (DDI) for fifty prostate volumetric modulated arc therapy plans were calculated, and compared to results predicted by machine learning using algorithms, namely, linear regression, tree regression, support vector machine (SVM) and Gaussian process regression (GPR). Root mean square error (RMSE), prediction speed and training time were determined to evaluate the performance of each algorithm. Results: From the results, it is found that the square exponential GPR algorithm had the smallest RMSE, relatively high prediction speed and short training time of 0.0038, 4,100 observation/s and 0.18 s, respectively. All linear regression, SVM and GPR algorithms performed well according to their RMSE in the range of 0.0038–0.0193. However, RMSE of the medium and coarse tree regression algorithms were found larger than 0.03, showing that they are not suitable for predicting DDI in this study. Conclusion: Machine learning can be used to predict dose-volume parameter such as DDI in radiation treatment planning QA. Selection of a suitable machine learning algorithm is important to determine the parameter effectively.

In this study, the effect of the core thickness of the sandwich composite on its flexural strength was investigated. E-glass fibre reinforced epoxy-based laminates were used as the skin materials whereas a polyurethane foam was used as the core material with varying thickness of 5, 7, and 9 mm. The experimental/theoretical and FEA analysis were carried out. Results indicated that as core thickness increased, the axial face stresses/core shear stresses and total deformation decreased. This improved the composite structural properties. This improvement was accompanied by a minor increase in weight. A good concurrence was observed between theoretical and FEA results. These results allow us to compare and select optimum parameters for aerospace applications.