Evolutionary, Neural and Fuzzy Systems for Smart and Micro Grids
1Politecnico di Milano, Milan, Italy
2Chouaib Doukkali University, El Jadida, Morocco
Evolutionary, Neural and Fuzzy Systems for Smart and Micro Grids
Description
The transition from low carbon to sustainable energy production is gaining momentum to support an increasing energy demand. This transition has led to an increase in the penetration of renewable energy sources in the power grid, which poses new challenges. In addition, the forecasted increase in the number of electric vehicles represents an increase in the variability of the electrical loads.
Computational intelligence (CI) provides a wide set of tools for effective designing, scheduling, and maintaining smart grids and microgrids. Among all CI techniques, a data-driven method based on machine learning (ML) that includes neural networks, random forests, deep learning, evolutionary optimization, and fuzzy logic has the potential to contribute to the modelling and analysis of future energy applications. Machine learning algorithms can be exploited for different time horizons which correspond to different decision-making activities. It can be applied in both production and load forecasting at various time scales. For example, deep learning algorithms can be employed on images or for time series analysis for improving the forecasting accuracy in very short-term forecasting.
Evolutionary optimization algorithms can handle nonlinear, multimodal, and high-dimensional problems in an effective way. EO can be used in the design phase for finding the optimal system configuration considering different objectives. It is possible to analyze the system stability, the operational and investment cost, the quality of service, and many others. These objectives can be combined to have a multiobjective problem. Finally, they can be used for system management directly or combined with fuzzy logic and ML techniques. Fuzzy logic can be implemented for the definition of advanced control logics for system management or even for maintenance purposes. They can effectively handle nonnumerical classifications that are very often provided by experienced operators.
The aim of this Special Issue is to bring together original research and review articles discussing this research topic. Submissions should focus on the application of computational intelligence techniques in smart and micro grids.
Potential topics include but are not limited to the following:
- Variable renewable energy sources forecasting with different time horizons (e.g., very short, short, intraday, and day-ahead)
- Load forecasting and electric vehicle charging time forecasting
- Advanced control strategies by means of computational intelligence for electric vehicle integration in smart and micro grids
- Fault diagnosis and analysis in electrical systems
- Evolutionary algorithms for sizing
- Evolutionary algorithms for optimal system design (e.g., single-objective and multi-objective analyses)
- Fuzzy and neuro-fuzzy applications to energy management and control of microgrid systems
- Optimization of renewable energy sources (e.g., solar photovoltaics, fuel cells, wind turbine and energy harvesting)
- Machine learning and computational intelligence applications for communication systems in smart and micro grids