Back to Search
Start Over
Parametric Studies on Artificial Intelligence Techniques for Battery SOC Management and Optimization of Renewable Power.
- Source :
- Procedia Computer Science; 2020, Vol. 167, p353-362, 10p
- Publication Year :
- 2020
-
Abstract
- Drastic increases in oil prices and need to reduce global warming are issues that have raised huge public awareness, which in turn encouraged many developed and developing countries to adopt new energy policies. Renewable energy resources are gaining fast attention in solving power supply problems in almost every area to shield the environment. In solving power supply problem standalone microgrid systems including renewable energy sources plays a key role. For a stand-alone renewable power delivery system, battery is the crucial component which allows storing excess renewable energy and further using the energy as and when needed by the load. In order to achieve the best performance and utilization the state of charge of the batteries should be assessed. This paper presents heuristic approaches to evaluate the state of charge (SOC) of lead–acid battery through optimal power generation in a standalone hybrid wind-solar renewable power delivery system. The mathematical analysis is made by the application of modified heuristic techniques, i.e. tuned genetic algorithm (TGA) and improved colliding body optimization(ICBO), for a specific loading condition in India. More optimized power is generated by the renewable distributed energy sources when evaluated by ICBO than that achieved by TGA, for both winter and summer load demands. These optimized value of power generated by the application of ICBO, resulted in low average value of SOC. The low values of SOC further prevent the battery from malfunctioning for a longer period of time. In order, to explore more optimal generation pertaining to good value of SOC parametric studies of modified algorithms, i.e. TGA and ICBO have been carried out. A set of critical operational parameters have also been obtained through parametric study of TGO and ICBO for a specific real time load profiles. Thus the reliable and optimal values lead to a better prospect in utilizing standalone renewable power delivery system for Indian future power scenario. Moreover, a comparative study has also been performed in analysing the performances of conventional and non-conventional methods in this regard. It has been found that the heuristic techniques, viz ICBO and TGA delivered 13% and 17% better result in comparison to that obtained by linear programming method respectively. Thus it can be surmise that the heuristic techniques perform well in considering the uncertainties and approximations of real world problem of generation renewable power, pertaining to better values of SOC. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 18770509
- Volume :
- 167
- Database :
- Supplemental Index
- Journal :
- Procedia Computer Science
- Publication Type :
- Academic Journal
- Accession number :
- 142768209
- Full Text :
- https://doi.org/10.1016/j.procs.2020.03.235