چكيده به لاتين
Today, computer science is widely used to solve various problems with all-round advances, especially in the field of artificial intelligence. Recent advances and new methods in the field of artificial intelligence has led to the solution of several problems in various fields of science and technology. One of the main issues in using artificial intelligence problem solving methods is the ability to coordinate and correctly understand the limitations and values of an issue.
Distributed artificial intelligence or the same as multi-agent systems (MAS), is one of the most widely used methods of problem solving in artificial intelligence. MAS are widely used to solve real-world problems due to their distributed nature, agent autonomy, failure resistance and other properties. A major challenge in using MAS is to properly analyze the problem and adapt it to a multi-agent system, which can cover all the limitations and goals of the system.
This thesis deals with a part of the smart grid problem that is in the residential field. In this field, which has been cited as an example, services should be timed to make the most of domestic energy resources and at the same time all services receive the energy they need.
In this regard, with a lot of study in related fields and research in previous works, an attempt has been made to optimize one of the previous works by carefully recognizing the limitations. In the initial work, unfavorable results were achieved in some circumstances. Then, by changing the simple function that had problems with valuation, defining a new, highly accurate, nonlinear "value" function, and considering all the changed conditions and constraints, the results tended to be optimized. Finally, this MAS, by adapting to different conditions, can be used while maintaining autonomy, confidentiality and tolerance to an acceptable level, and also by using more local resources, reduce material and environmental costs.
