Drought is the least understood natural disaster due to the complex relationship of multiple contributory factors. Its beginning and end are hard to gauge, and they can last for months or even for years. India has faced many droughts in the last few decades. Predicting future droughts is vital for framing drought management plans to sustain natural resources. The data-driven modelling for forecasting the metrological time series prediction is becoming more powerful and flexible with computational intelligence techniques. Machine learning (ML) techniques have demonstrated success in the drought prediction process and are becoming popular to predict the weather, especially the minimum temperature using backpropagation algorithms. The favourite ML techniques for weather forecasting include support vector machines (SVM), support vector regression, random forest, decision tree, logistic regression, Naive Bayes, linear regression, gradient boosting tree, k-nearest neighbours (KNN), the adaptive neuro-fuzzy inference system, the feed-forward neural networks, Markovian chain, Bayesian network, hidden Markov models, and autoregressive moving averages, evolutionary algorithms, deep learning and many more. This paper presents a recent review of the literature using ML in drought prediction, the drought indices, dataset, and performance metrics.

This paper aims to evaluate the performance of multiple non-linear regression techniques, such as support-vector regression (SVR), k-nearest neighbor (KNN), Random Forest Regressor, Gradient Boosting, and XGBOOST for COVID-19 reproduction rate prediction and to study the impact of feature selection algorithms and hyperparameter tuning on prediction. Sixteen features (for example, Total_cases_per_million and Total_deaths_per_million) related to significant factors, such as testing, death, positivity rate, active cases, stringency index, and population density are considered for the COVID-19 reproduction rate prediction. These 16 features are ranked using Random Forest, Gradient Boosting, and XGBOOST feature selection algorithms. Seven features are selected from the 16 features according to the ranks assigned by most of the above mentioned feature-selection algorithms. Predictions by historical statistical models are based solely on the predicted feature and the assumption that future instances resemble past occurrences. However, techniques, such as Random Forest, XGBOOST, Gradient Boosting, KNN, and SVR considered the influence of other significant features for predicting the result. The performance of reproduction rate prediction is measured by mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), R-Squared, relative absolute error (RAE), and root relative squared error (RRSE) metrics. The performances of algorithms with and without feature selection are similar, but a remarkable difference is seen with hyperparameter tuning. The results suggest that the reproduction rate is highly dependent on many features, and the prediction should not be based solely upon past values. In the case without hyperparameter tuning, the minimum value of RAE is 0.117315935 with feature selection and 0.0968989 without feature selection, respectively. The KNN attains a low MAE value of 0.0008 and performs well without feature selection and with hyperparameter tuning. The results show that predictions performed using all features and hyperparameter tuning is more accurate than predictions performed using selected features.

No abstract

Powder bed fusion (PBF) applies to various metallic materials used in the metal printing process of building a wide range of complex parts compared to other AM technologies. PBF process has several variants such as DMLS (direct metal laser sintering), EBM (electron beam melting), SHS (selective heat sintering), SLM (selective laser melting), and SLS (selective laser sintering). For PBF to reach its maximum potential, machine learning (ML) algorithms are used with suitable materials to achieve goals cost-effectively. Various applications of neural networks, including ANNs, CNNs, RNNs, and other popular techniques such as KNN, SVM, and GP were reviewed, and future challenges were discussed. Some special-purpose algorithms were listed as follows: GAN, SeDANN, SCNN, K-means, PCA, etc. This review presents the evolution, current status, challenges, and prospects of these technologies in terms of material, features, process parameters, applications, advantages, disadvantages, etc., to explain their significance and provide an in-depth understanding of the same.

The friction welding of tube to tube plate using an external tool (FWTPET) is widely deployed in several industrial applications, such as aerospace, automotive, and power plants. Moreover, for achieving a better tensile strength and hardness in the weld zone, the friction stir processing (FSP) technique was incorporated into the FWTPET process for joining aluminum alloys (AA6063 tube, AA6061 tube plate). Furthermore, it has to be noted that FWTPET was applied for joining the AA6063 tube to the AA6061 tube plate, and FSP was deployed for reinforcing the weld zone with carbon nanotube (CNT) and silicon nitride (Si3N4) particles, thereby attaining the desirable mechanical properties. Subsequently, the Taguchi L25 orthogonal array was used for identifying the most influential input and output FWTPET + FSP process parameters. Furthermore, particle swarm optimization (PSO) and the firefly algorithm (FFA) were deployed for determining the optimized input and output FWTPET + FSP process parameters. The input process parameters include CNT, Si3N4, rotational tool speed, and depth. Furthermore, the tensile strength of the welded joint was considered as the output process parameter. The process parameters predicted by PSO and FFA were compared with the experimental values. It was witnessed that deviation between the predicted and experimental values was minimal. Moreover, it was found that FFA provided a superior tensile strength prediction than PSO.