A Heterogeneous Multi-Agent System Model With Navigational Feedback for Load Demand Management of a Zonal Medium Voltage DC Shipboard Power System

Increased demand of electric ship power with emerging requirements for serving highly dynamic loads at limited power sources, has motivated the development of medium voltage DC shipboard power systems. As different types of power converters can be involved in the same system, advanced load management scheme is required to ensure stable and optimal operation under various conditions. In this paper, a heterogeneous multi-agent system model is established for the load demand management of a zonal medium voltage DC shipboard power system, where different types of DC-DC converters with diverse conversion ratios are considered as heterogeneous agents. Then, a flocking based cooperative control protocol is designed to achieve the group objective, where navigational feedback is introduced to ensure that each agent runs within a limited range. Finally, numerical simulations are conducted to verify the performance of the proposed model. Compared with the existing work, our contribution lies in: 1) The heterogeneous multi-agent system model we formulated for a combination of DC-DC converters with different types and diverse voltage levels is more consistent with the actual power system and each converter can be controlled individually according to load ratings and priorities; 2) The modified navigational feedback with the function of confining the lower and upper bound of agents can significantly improve the control effect of the cooperative control protocol.