Sustainability in Energy Economy and Environment: Role of AI Based Techniques

Abstract

With the ever increasing electricity demand in the twenty-first century, conventional power grids have to respond by adding new generation capacities. This, however, requires upgradation of transmission and distribution systems over utility structures in an environmentally sustainable manner, further expanding to new remote/rural areas. Thus, to retain techno-economical and environmental viability, the concept of decentralized power delivery systems has come up as a new alternative. Though the search for technological effectiveness of microgrids is important, the establishment of a standardized and competitive price structure in the electricity market is also required. Hence, nowadays, economic evaluation of decentralized power delivery systems are being dealt with proper attention. Most conventional methods are classical algorithms, which generally include replication and precision with a lengthy and extensive search process. These deterministic methods involve trial and error process, without any guidance, hint, adaptation and self-correction. In order to overcome such problems modern optimization techniques have emerged which do a random search on basis of some plausible hints and find solution by trial and error. Though these elementary soft-computing methods can attain possible solutions within a given finite time, they never assure to achieve the best solution. Thus to achieve better and more accurate solutions, artificial intelligence techniques with attributes such as partial truth, uncertainties, imprecise specification and approximation which are often present in problems of practical relevance in energy economics, will be explored. Further, considering the relative strength and weakness of different soft-computing techniques, hybrid algorithms will be discussed and assessed in terms of performance this chapter.