Abstract
Global climate changes and the increase in average temperatures are some of the major contemporary problems that have not been considered in the context of external factors to increase accident risk. Studies that include climate information as a safety parameter in machine learning models designed to predict the occurrence of accidents are not usual. This study aims to create a dataset with the most relevant climatic elements, to get better predictions. The results will be applied in future studies to correlate with the accident history in a retail sector company to understand its impact on accident risk. The information was collected from the National Oceanic and Atmospheric Administration (NOAA) climate database and computed by a wrapper method to ensure the selection of the most features. The main goal is to retain all the features in the dataset without causing significant negative impacts on the prediction score. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract

Abstract
[No abstract available]

Abstract

Abstract
To ensure the autonomous power supply in microgrids (MGs) in stand-alone mode while also maintaining stability, energy storage systems (ESSs) and demand-side flexibility can be utilized together. Motivated by this fact, in this study, a scenario-based energy management system (EMS) modelled as a mixed-integer linear programming (MILP) problem is presented by taking the stochastic nature of wind and photovoltaic (PV) sources into account in order to analyze the operational behaviour of MGs and thereby to reduce the network energy losses. Direct load control (DLC) based demand response (DR) program is implemented to the system with the objective of exploiting the remarkable potential of thermostatically controllable appliances (TCAs) for energy reduction while satisfying comfort and operational constraints. Furthermore, a common ESS with a bi-directional power flow facility is incorporated in the proposed structure and electric vehicles (EVs) are employed as an additional flexible load in grid-to-vehicle (G2V) mode. To testify the effectiveness of the proposed optimization algorithm, different case studies are conducted considering diverse scenarios. Moreover, the performance is compared with a deterministic method from the perspective of achieving loss reduction and capturing the uncertainties.

Abstract
In recent years, with the growth of digital media and modern imaging equipment, the use of video processing algorithms and semantic film and image management has expanded. The usage of different video datasets in training artificial intelligence algorithms is also rapidly expanding in various fields. Due to the high volume of information in a video, its processing is still expensive for most hardware systems, mainly in terms of its required runtime and memory. Hence, the optimal selection of keyframes to minimize redundant information in video processing systems has become noteworthy in facilitating this problem. Eliminating some frames can simultaneously reduce the required computational load, hardware cost, memory and processing time of intelligent video-based systems. Based on the aforementioned reasons, this research proposes a method for selecting keyframes and adaptive cropping input video for human action recognition (HAR) systems. The proposed method combines edge detection, simple difference, adaptive thresholding and 1D and 2D average filter algorithms in a hierarchical method. Some HAR methods are trained with videos processed by the proposed method to assess its efficiency. The results demonstrate that the application of the proposed method increases the accuracy of the HAR system by up to 3% compared to random image selection and cropping methods. Additionally, for most cases, the proposed method reduces the training time of the used machine learning algorithm.