Optimal sizing of a multi-energy system using a multi-agent decentralized operation model considering private-ownership

Abstract

In this study, the sizing of a multi-energy system (MES) based on a multi-agent system (MAS) has been illustrated. In this model, the operation of the MES is completely decentralized and the sizing of the system has been done by adding an agent which takes the responsibility of the sizing optimization. The remarkable point is that each agent calculates the Net Present Cost (NPC) of its entities locally by getting the size of the elements from the optimizer agent and prevent agents from disclosing their private data. Then the optimizer agent updates the optimal size of elements in the multi-energy system by the calculated Net Present Cost which receives from other agents and selects the best answer from the available optimal answers based on the social welfare index. Our simulation result shows that the NPC of −196,161.88 $ is the best answer based on the assumed specifications. Given the fact that in the process of obtaining Net Present Costs the central agent is not involved and each agent gets the optimal Net Present Cost only by direct communication with its neighbor, this model eases the private ownership implementation in the sizing of the MESs.