Smart cities are one of the promising application areas where the energy management (supply/dispatch) issues have the potential to positively affect the society. In the common era of information and technology, the consumers expect the required amount of energy to be present at anytime. This energy is obtained through renewable and nonrenewable resources. Smart wind energy is the idea of efficient energy generation utilizing the wind while satisfying green expectations. However, the random characteristics caused by various external factors have significant effects on wind speeds, thus introducing difficulties in power systems operations and energy generation. Forecasting techniques can be effectively used to predict future wind speeds and power generation to optimize the energy output. The objective of this study is to provide a framework for a local smart-city wind energy harvesting model. Our study evaluates wind speed forecasting models based on synced weather characteristics (air temperature, humidity, and pressure), against models based on wind speed data itself. The produced models are implemented via k-nearest neighbors and linear regression. Two different data sets are addressed for performance comparison purposes. The first one is 5 years of 10-minute time span measurements at Middle East Technical University campus. The second data set is 3 years of the same time span at a site in California, USA. The accuracy of predictions is found to be higher for the data sets with lower variance (METU data set). However, the models fit the higher variance (NREL data set) data better than the lower one.
In this paper, we provide a guideline for using the Neural Network Dependability Kit (NNDK) during the development process of NN models, and show how the algorithm is applied in two image classification use cases. The case studies demonstrate the usage of the dependability kit to obtain insights about the NN model and how they informed the development process of the neural network model. After interpreting neural networks via the different metrics available in the NNDK, the developers were able to increase the NNs' accuracy, trust the developed networks, and make them more robust. In addition, we obtained a novel application-oriented technique to provide supporting evidence for an NN's classification result to the user. In the medical image classification use case, it was used to retrieve case images from the training dataset that were similar to the current patient's image and could therefore act as a support for the NN model's decision and aid doctors in interpreting the results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.