Energy Resilience and Climate Protection by Heinz-Adalbert Krebs & Patricia Hagenweiler

Author:Heinz-Adalbert Krebs & Patricia Hagenweiler
Language: eng
Format: epub
ISBN: 9783658375645
Publisher: Springer Fachmedien Wiesbaden

5.6.1 Artificial Intelligence

Artificial intelligence (AI) is developing as a basic innovation to drive digitization and autonomous systems in all areas of life in our society, economy, administration and science. The progress of artificial intelligence (AI), especially in the field of machine learning, is primarily due to the exponential increase in hardware performance and its use for processing large data sets.38 Artificial intelligence (AI) compresses, among other things, the use of machine learning, natural language processing, and computer vision, whereby computers learn to make predictions and recognize patterns, such as in the context of self-driving (automated) vehicles, which can recognize objects and predict driving behavior. In the context of artificial intelligence, large amounts of data are required and their quality is essential, as biased and inaccurate data can lead to errors in predictions and pattern recognition. In addition, the data-driven network economy is leading to a market concentration of Big Data (huge amounts of data) by a few large companies in North America, the European Union (EU), and China, which account for more than 90% of patent applications and raising risk capital in artificial intelligence (AI). Last but not least, the value of data has increased, impacting privacy, cyber security, and surveillance. As pandemics and other crises unfold, there is a need to build large public databases on which artificial intelligence can be trained to help with prediction, tracking, and diagnostics, especially as containment strategies based on better data analytics should enable the resumption of economic activity and prevent further waves of infection.39

Artificial intelligence (AI) can be applied to many different risks and used at all stages of the disaster management cycle. For example, AI models are already being applied to dynamic risk analysis, which can help identify vulnerabilities, detect risks early, and predict their evolution. These include, for example, regular local forecasts of flood and landslide risk, which can be used to issue early and location-specific safety warnings. In the field of critical infrastructures, such as smart grids, artificial intelligence (AI) applications are increasingly being used for the decentralized control and optimization of microgrids, the classification of the type and severity of grid failures, and the forecasting of electricity demand, electricity prices and electricity generation by photovoltaic and wind power plants.40

In the context of the Corona pandemic, numerous new applications using artificial intelligence have emerged, but their usefulness cannot yet be assessed, especially since the artificial intelligence (AI) currently in use is based on statistical learning from large data sets. Against this background, the ability of artificial intelligence is still limited in that learning abstract concepts from a small number of examples, understanding causality, transferring what has been learned to other domains, and dealing with hierarchical structures is not yet possible. Without the presence of appropriate historical data or virtual training environments, AI systems are not yet able to produce a good result. For example, the accuracy and resolution of demand forecasting in the power grid depends on the availability of granular data from smart meters. In particular, for use

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.