computational intelligence

As power demand rises, the power system becomes more stressed, potentially leading to an increase in power losses. When compared to lower power losses, higher power losses result in higher power system operating cost. Flexible AC Transmission System (FACTS) devices help to reduce power losses. This paper describes the use of a computational intelligence-based technique, in this case the Artificial Immune System (AIS), to solve the installation of Thyristor Controlled Static Compensator (TCSC) and Static VAR Compensator (SVC) in a power system while ensuring optimal sizing of both devices. The goal of determining the best locations and sizes for the multi-type FACTS devices is to minimize system power loss. Three case studies are presented to investigate the effectiveness of the proposed AIS optimization technique in solving the multi-type FACTS device installation problem under various power system conditions. The optimization results generated by the proposed AIS are beneficial in improving the power system, particularly in terms of system power loss minimization, which also contributes to power system operating cost minimization. As a result, the likelihood of this being sustainable and able to be implemented for an extended period is greater.

With the rapid expansion of the information on the Internet, recommender systems play an important role in terms of trade and research. Recommender systems try to guess the user's way of thinking, using the in-formation of user's behavior or similar users and their views, to discover and then propose a product which is the most appropriate and closest product of user's interest. In the past decades, many studies have been done in the field of recommender systems, most of which have focused on designing new recommender algorithms based on computational intelligence algorithms. The success of a recommender system besides the quality of the algorithm depends on other factors such as: Sparsity, Cold start and Scalability in the performance of a recommender system, which can affect the quality of the recommendation. Consequently, the main motivation for this research is to providing an effective meta heuristic algorithm based on a combination of imperialist competitive and firefly algorithms using clustering technique. The simulation results of the proposed algorithm on real data sets Move Lens and Film Trust have shown better forecast accuracy in the item recommendation to users than other algorithms presented in subject literature. Also the proposed algorithm can choose appropriate items among the wide range of data and give it to output in a reasonable time.